|Home | About | Journals | Submit | Contact Us | Français|
CD1d-restricted invariant NKT (iNKT) cells are important immunoregulatory cells in antitumor immune responses. However, the quantitative and qualitative defects of iNKT cells in advanced multiple myeloma (MM) hampered their antitumor effects. Therefore, the development of functional iNKT cells may provide a novel strategy for the immunotherapy in MM treatment.
We activated and expanded iNKT cells from MM patients with α-galactosylceramide(α-GalCer)-pulsed-dendritic cells (DCs), characterized their antitumor effects by the cytokine production profile and cytotoxicity against MM cells, and explored the effects of immunomodulatory drug lenalidomide on these iNKT cells. We also investigated the expression of CD1d by primary MM cells and its function to activate iNKT cells.
We established highly purified functional iNKT cell lines from newly diagnosed and advanced MM patients. These CD1d-restricted iNKT cell lines produced high level of antitumor Th1 cytokine in response to α-GalCer-pulsed-primary MM cells, CD1d-transfected MM1S cell line or DCs. Moreover, MM iNKT cell lines displayed strong cytotoxicity against α-GalCer-pulsed-primary MM cells. Importantly, lenalidomide further augmented the Th1-polarization by iNKT cell lines via the increased Th1 cytokine production and the reduced Th2 cytokine production. We also demonstrated that CD1d was expressed in primary MM cells at mRNA and protein levels from the majority of MM patients, but not in normal plasma cells and MM cell lines, and CD1d+ primary MM cells presented antigens to activate iNKT cell lines.
Taken together, our results provide the pre-clinical evidence for the iNKT cells-mediated immunotherapy and a rationale for their use in combination with lenalidomide in MM treatment.
Multiple myeloma (MM) is still a fatal hematological malignancy characterized by the accumulation of terminally differentiated plasma cells in the bone marrow of patients (1). Although high-dose chemotherapy with stem cell transplantation has shown some success, the outcome of the majority of patients with MM is unsatisfactory (2). Clinical benefits may be obtained from immunotherapy to stabilize or even eradicate minimal residual disease after the conventional treatments for patients with MM.
Invariant natural killer T cells (iNKT cells) constitute an innate lymphocyte lineage that has an important role in regulating immune responses, including antitumor responses. iNKT cells display an extremely restricted T cell antigen receptor (TCR) repertoire, in humans consisting of a specific Vα24-Jα18 chain rearrangement preferentially paired with a Vβ11 chain. Unlike conventional T cells that recognize peptide antigens, iNKT cells recognize glycolipid ligands presented by a non-polymorphic MHC class 1-like antigen presenting molecule CD1d, and are characterized by their capacity to rapidly produce large amounts of immunoregulatory cytokines (3). iNKT cells play a physiologic role in tumor immunosurveillance against carcinogen-induced tumors (4) and are required for the antitumor effects of low-dose interleukine (IL)-12 treatment (5). Most importantly, pre-clinical study in murine models have demonstrated that, upon activation by α-galactosylceramide (α-GalCer), a highly specific ligand for CD1d, iNKT cells can stimulate potent antitumor immune responses through the production of Th1 cytokines (6–8). iNKT cells have also shown the directed killing activity against CD1d+ tumor cells (9–11).
In progressive multiple myeloma, however, iNKT cells are functionally defective evidenced by deficient ligand-dependent IFN-γ production, resulting in a detrimental Th2 cytokine profile (12). Similarly, studies from several groups indicate that iNKT cells are decreased and/or functionally impaired in various cancer patients including prostate cancer, melanoma and myelodysplastic syndromes (13–16). Therefore, a novel immunotherapeutic strategy may be developed using iNKT cell adoptive transfer to MM patients after in vitro expansion and functional activation.
Lenalidomide (CC-5013, Revlimid, and IMiD3), which belongs to a class of thalidomide analogs known as the immunomodulatory drugs, was approved for the treatment of MM in 2006. Lenalidomide induces apoptosis, decreases the binding of myeloma cells to stromal cells in bone marrow and inhibits angiogenesis (17). Additionally, lenalidomide increases conventional T cell costimulation and NK cell cytotoxicity (18, 19), and a report has recently revealed that lenalidomide can also enhance ligand-dependent activation of iNKT cells (20).
In this study, we evaluated the expression and function of CD1d on MM tumor cells. We established iNKT cell lines from MM patients, characterized their antitumor profile, and further addressed the effects of lenalidomide on these CD1d-restricted iNKT cells. Our research provides the preclinical basis and rationale for the use of iNKT cells in antimyeloma immunotherapy.
Healthy donor leukopacks were obtained from Dana Farber Cancer Institute; Normal bone marrow samples were purchased from AllCells, LLC (CA); MM patient blood and bone marrow samples were obtained from Dana Farber Cancer Institute and Veterans Administration (VA) Boston Healthcare System following informed consent approved by the institutional review boards. Patients were classified as multiple myeloma, according to standard diagnostic criteria.
DCs were generated from adherent peripheral blood mononuclear cells (PBMCs) in the presence of GM-CSF (1000U/mL) and IL-4 (25ng/mL, both from R&D system, Minneapolis, MN). On day 5~6, the DCs were matured by LPS (100ng/mL; Sigma-Aldrich) overnight in the presence of 100ng/mL of α-GalCer (KRN7000, Kirin Brewery, Gunma, Japan) and irradiated at 50Gy immediately before use.
Due to the low frequency of iNKT cells in PBMCs, iNKT cells were first enriched by staining with unconjugated anti-TCRVα24 mAb (Immunotech, Marseille, France) followed by the immunomagnetic isolation with goat anti-mouse IgG microbeads (Miltenyi Biotec, Auburn, CA). Subsequently, Vα24+ cells and α-GalCer-pulsed mature DCs were cocultured in PRMI 1640 medium supplemented with 10% of heat-inactivated fetal bovine serum (FBS), 15mM HEPES, 5.5 ×10−5 M β-mercaptoethanol, 50 units/mL penicillin, 50 μg/mL streptomycin, and 2 mmol/L L-glutamine. Recombinant human IL-2 (100U/mL, Chiron Corporation, CA) was added to the cultures on day 2 and supplemented every 4~6 days. When the cells had proliferated, iNKT cells were further enriched with PE-conjugated anti-TCRVβ11 (Immunotech) mAb followed by immunomagnetic isolation with anti-PE microbeads (Miltenyi Biotec). Culture was then gradually expanded and restimulated every 7~10 days. The frequencies of iNKT cells pre- or post- expansion were monitored by quantifying Vα24+ Vβ11+ cells using flow cytometric analysis. Considering the low frequency of iNKT cells in PBMCs, at least 1~3×105 cells were collected for each analysis in the lymphocyte gates.
Normal plasma cells and primary MM cells were purified from bone marrow mononuclear cells by positive selection with CD138 microbeads (Miltenyi Biotec), according to the manufacturer’s instructions.
Gene expression of CD1d was performed by microarray analysis. Total RNA from normal plasma cell samples, MM cell lines and primary MM cells was isolated utilizing an “RNeasy” kit (Qiagen Inc., USA) and gene expression profile was evaluated using HG-U133 arrays (Affymetrix, Santa Clara, CA). GeneChip arrays were scanned on a GeneArray Scanner (Affymetrix). Array normalization, expression value calculation and clustering analysis were performed using the dChip Analyzer.
Multiple myeloma cell line MM1S (kindly provided by Dr. Steven Rosen, Northwestern University, Chicago, IL) transfected with a CD1d cDNA in the pSRα-neo expression vector (21) and empty vector using Nucleofector Kit V (Amaxa Biosystems, Cologne, Germany) according to the manufacturer’s instructions. CD1d+ MM1S cells were further isolated with magnetic microbeads and followed by selection with G418 for generating the cell line stably expressing CD1d. Mock cell line was only selected by G418. CD1d expression was determined by immunofluorescence using anti-CD1d-PE mAb. The expression of CD138 on transfected MM1S cells was detected by anti-PE-CD138-PE mAb to monitor the change of MM1S cell line. CD1d- and mock- transfected MM1S cell lines were grown in RPMI 1640 supplemented with 10% FBS, 50 units/mL penicillin, 50 μg/mL streptomycin, and 2 mmol/L L-glutamine.
Mock/CD1d-transfected MM1S cells and matured DCs were pre-loaded with vehicle or α-GalCer (100ng/mL) overnight. The cells were then washed to remove the supernatant before use. 1 × 105 rested iNKT cells from MM patients were added into 96-well plate with above mock/CD1d-transfected MM1S cells or DCs at ratio of 2:1 and 4:1 respectively. Supernatants were collected at 48h for IL-2 and IL-4, and 72h for IFN-γ measurements, respectively.
1 × 105 rested iNKT cells were plated in 96-well plates with either medium alone or with 5×104 primary MM cells. α-GalCer (100 ng/mL) was added as indicated. IL-2 was included in the culture medium at a final concentration of 10 units/mL. The supernatants were collected for the cytokine detection at 48 hours. Meanwhile, the cells were harvested and stained by anti-TCRVα 24-FITC and anti-CD25-PC5 mAbs.
Lenalidomide (Celgene, Warren, NJ) was dissolved in dimethyl sulfoxide (DMSO) and stored at −20°C. Drug was diluted in culture medium with less than 0.1% DMSO immediately before use. The following systems were set up: iNKT cells alone, iNKT cells cultured with MM1S.CD1d in the absence or presence of α-GalCel (100 ng/mL), with or without the treatment of lenalidomide (2μM). The supernatant was collected at 72 hours and subject to ELISA detection.
Released cytokine levels were determined by ELISA. IFN-γ, IL-2 and IL-4 were quantified by the Quantikine immunoassay (R&D Systems, Minneapolis, MN) according to the instruction of the manufactures.
Cytotoxicity was assessed using a 4-hour calcein-AM- release assay as described previously (22). MM1S.CD1d cells or primary MM cells were labeled with calcein-AM (5μg/ml, Molecular Probe Inc, CA) and used as target cells at 5000 cells/well, in the presence or absence of α-GalCer (100 ng/mL). iNKT cell lines from healthy donors and MM patients were used as effector cells. The ratio of effector cells and target cells was 20:1. Calculation of cytotoxicity was done using the following equation: % cytotoxicity = 100 × (experimental release-spontaneous release)/(maximal release- spontaneous release)
The following antibodies were used for the flow cytometric analysis: FITC- or PE -conjugated anti-TCRVα24, anti-TCRVβ11, anti-CD16, anti-CD161 mAbs (Immunotech, Marseille, France). FITC-, PE- or PC5-conjugated anti-CD1d, anti-CD4, anti-CD8, anti-CD25, anti-CD56 and anti-CD94 mAbs (BD pharmingen, San Diego, CA). PE-conjugated anti-CD138 mAb (Miltenyi Biotec). Isotype-matched mAbs were used as controls. Flow cytometry was performed with a Coulter® EPICS® Elite flow cytometer (Beckman-Coulter).
Statistics analyses were performed by student t test. P < 0.05 was considered significant.
To develop an immunotherapy approach using iNKT cells, we first determined the frequency of iNKT cells in the PBMCs from healthy donors and MM patients. Consistence with published data in other types of cancer, the frequency of iNKT cells was reduced in patients with advanced MM (0.01±0.008%; n=7) compared to healthy controls (0.064±0.030%; n=7) (p<0.001). We established primary iNKT cell lines from healthy donors, newly diagnosed and advanced MM patients. Healthy donors exhibited a near 100% success rate for iNKT cell line generation, whereas MM patients exhibited a 50% success rate (5 out of 10 patients). Flowcytometry demonstrated greater than 97% purity of Vα24+Vβ11+ cells (Figure 1). The phenotype analysis on MM iNKT cell lines showed majority cells were CD4+ or CD4−CD8−. The expression of CD161 was variable, while no iNKT cell lines expressed CD16 or CD94. A certain level of CD56 expression was also observed (table 1). No significant phenotypic difference observed between established iNKT cell lines from healthy donors and MM patients. Generally, greater than 108 iNKT cells could be harvested in 6~8 weeks.
To evaluate potential in vivo interaction between iNKT cells and myeloma cells, we studied the profile of CD1d expression on primary MM cells as well as MM cell lines by gene expression profiling. Total 15 CD138+ primary MM cell samples and 6 MM cell lines (MM1S, ARD, ARK, ARP, PRMI8226, and U266) were compared to normal plasma cells; the majority of primary MM cells expressed higher levels of CD1d (11 out of 15). In contrast, all 6 MM cell lines tested shown no expression of CD1d (figure 2a). Flow cytometric analysis using anti-CD1d-PE mAb further confirmed the expression of CD1d on primary MM cells but lack expression on 12 different MM cell lines (MM1S, MM1R, ARD, ARK, ARP, PRMI8226, U266, OPM1, OPM2, CAG, 12PE and 28PE) (figure 2b and data not shown).
Since CD1d is expressed by primary myeloma cells but not by MM cell lines, we therefore established a stable CD1d-transfected MM1S cell line (MM1S.CD1d) for the feasibility of the functional study (Figure 2c). Flow-cytometry showed that 100% of the transfected cells expressed CD138. No Phenotypic change or growth characteristic difference was observed in the CD1d transfected MM1S cells compared to the parental cell line (data not shown).
To confirm the function and CD1d-reactivity of α-GalCer-expanded iNKT cells from MM, we evaluated their cytokine profile using mock/CD1d-transfected MM1S cells and DCs. iNKT cell lines stimulated by mock-transfected MM.1S cells produced very low to undetectable levels of IL-2, IFN-γ and IL-4, with or without α-GalCer (data not shown). When cultured with MM1S.CD1d cells alone, iNKT cells secreted low levels of these cytokines. Notably, when stimulated with α-GalCer-pulsed MM1S.CD1d cells, iNKT cells produced high levels of IFN-γ as well as IL-2, and a low amount of IL-4 (figure 3). These data were confirmed using α-GalCer-pulsed DCs, which resulted in a more pronounced Th1 immune response (Figure 3). In addition, we observed IL-10 production was at the low level when iNKT cell lines responded to MM1S.CD1d cells, in the presence or absence of α-GalCer. Together our results indicate that α-GalCer-reactive CD1d-restricted iNKT cell lines can be obtained from MM patients and exhibited a Th1 antitumor cytokine profile.
We further addressed whether iNKT cell lines activation could be mediated by CD1d+ primary MM cells. As shown in Figure 4, CD25 expression and IFN-γ production by iNKT cells were dramatically increased in the presence of α-GalCer-pulsed primary MM cell, demonstrating the efficiency of CD1d-mediated antigen presentation by MM tumor cells. Moreover, the functional cytokine profile evaluated by IFN-γ and IL-4 production further confirmed the Th1 anti-tumor iNKT cell lines obtained. In addition, significant increases in CD25 expression and IFN-γ production also observed when iNKT cells cocultured with primary MM cells in the absence of α-GalCer.
To further augment iNKT cell immune responses, we evaluated the effect of lenalidomide, an immunomodulatory drug, on iNKT cell lines. Lenalidomide did not directly stimulate iNKT cells (data not shown), but significantly increased IFN-γ and IL-2 production, and significantly decreased IL-4 production when iNKT cells were activated using α-GalCer-pulsed-MM1S.CD1d cells (Figure 5) and CD1d-transfected C1R cells (data not shown). Furthermore, lenalidomide also increased IFN-γ and IL-2 production when iNKT cells were cultured with MM1S.CD1d cells in the absence of α-GalCer (Figure 5).
We also investigated whether the expanded iNKT cell lines had the direct killing activity against MM cells. MM1S.CD1d cells and CD1d+ primary MM cells were used as target cells. The iNKT cell lines from both healthy donors and MM patients showed strong cytotoxicity against α-GalCer-loaded target cells. Meanwhile, a low level of cytotoxicity was found in the absence of α-GalCer (Figure 6). No cytotoxicity observed by iNKT cell lines against CD1d-negative primary MM cells or MM1S mock cells in the absence or presence of α-GalCer (data not shown).
Various studies have demonstrated an important role of iNKT cell-derived Th1 type cytokines in initiating antitumor immune responses. Through the production of IFN-γ, iNKT cells can stimulate the activation of downstream effectors including T cells, NK cells, DCs and macrophages and increase NK and T cell proliferation and cytotoxicity through IL-2 production (23–28). However, both the quantitative and qualitative defects of iNKT cells in advanced MM hampered their antitumor effects. In this study, we developed a novel immunotherapeutic strategy directed at the activation and expansion of Th1-polarized iNKT cells from MM patients. We report the establishment of CD1d-restricted iNKT cell lines from newly diagnosed and advanced MM patients. These iNKT cell lines produced Th1 cytokines, indicating that the suppressive effects of the MM microenvironment on iNKT cells have been overcome by in vitro culture. Although MM patients have a very low frequency of iNKT cells, combination of antibodies purification and selective stimulation with α-GalCer-pulsed DCs, allowed us to obtain high yields iNKT cell lines, which provided adequate cell numbers for the potential application of adoptive immunotherapy.
We further demonstrated that CD1d was expressed by tumor cells in majority of MM patients. This phenomenon indicates an ideal environment for iNKT cell-mediated immunotherapy in MM. First, the CD1d+ MM cells provide the direct targets for iNKT cells. Our data have shown the strong killing activity by iNKT cell lines against α-GalCer pulsed- MM cells. Moreover, CD1d molecule on patient MM cells effectively present α-GalCer to activate iNKT cells, resulting in the Th1-polarized cytokine production. Therefore, another immunotherapeutic strategy may be developed using an autologous tumor cell-based vaccine consisting of irradiated α-GalCer-pulsed CD1d+ myeloma cells to boost iNKT cells activity in vivo. We also revealed that iNKT cells displayed a certain level of reactivity when culture with CD1d+ MM cells, even in the absence of exogenous α-GalCer, suggesting that MM cells may express natural CD1d-bound ligands for iNKT cells. Further identification of these ligands would provide a basis for generation of MM-specific iNKT cells.
Importantly, we demonstrated that immunomodulatory drug lenalidomide further augments the antitumor effects of iNKT cell lines via CD1d-restricted stimulation, evidenced by not only enhancing the Th1 cytokine production, but also reducing the Th2 cytokine production. Our results provide the rationale for the combination with lenalidomide in iNKT cell-mediated therapy. Currently the mechanism of lenalidomide on iNKT cells is unclear. For conventional T cells, our group revealed that lenalidomide via the B7-CD28 pathway to co-stimulate these cells (18). We have detected the CD28 expression on our iNKT cell lines from healthy donors and MM patients. It showed all the iNKT cell lines express CD28 (data not shown). The further characterization of whether CD28 and/or other costimulatory signaling pathways involved will delineate mechanism of the effect by lenalidomide upon CD1d restricted activation.
In summary, our study demonstrates that functional iNKT cell lines can be generated from MM patients with α-GalCer-pulsed DCs and further improved by lenalidomide. These results provide the preclinical feasibility and rationale for iNKT cell-mediated immunotherapy in MM.
Grant Support: Supported in part by National Institutes of Health grants P050-100707 and PO1-78378 (N.C.M and K.C.A), and a Merit Review Award from the Research Service Veterans Health Care (N.C.M). N.C.M. is a Leukemia Society Scholar in Translational Research.
We thank Kirin Brewery Co. Ltd. (Gunma, Japan) for kindly providing α-galactosylceramide (α-GalCer).