It remains unclear how the immune system affects leukemia development. To clarify the significance of the presence of immune systems in leukemia development, we transferred MLL/ENL leukemia cells into immune-competent or immune-deficient mice without any preconditioning including irradiation. The wild-type mice did not develop leukemia, whereas all the Rag2−/−γc−/− mice lacking both adaptive immune cells and natural killer (NK) cells developed leukemia, indicating that leukemia cells were immunologically rejected. Interestingly, leukemia cells were also rejected in 60% of the Rag2−/− mice that lacked adaptive immune cells but possessed NK cells, suggesting that NK cells play a substantial role in the rejection of leukemia. Moreover, engraftment of leukemia cells was enhanced by NK cell depletion in Rag2−/− recipients and inhibited by transfer of NK cells into Rag2−/−γc−/− recipients. Upregulation of NKG2D (NK group 2, member D) ligands in MLL/ENL leukemia cells caused elimination of leukemia cells by NK cells. Finally, we found that leukemia cells resistant to elimination by NK cells had been selected during leukemia development in Rag2−/− recipients. These results demonstrate that NK cells can eradicate MLL/ENL leukemia cells in vivo in the absence of adaptive immunity, thus suggesting that NK cells can play a potent role in immunosurveillance against leukemia.
The cancer immunosurveillance hypothesis predicts that tumor cells are recognized by the immune system, leading to the subsequent eradication of tumors.1 In addition, accumulating evidence indicates that the immune system not only eliminates developing tumors but also sculpts the immunogenicity of tumor cells and finally generates a tumor cell repertoire that is capable of survival in immune-competent hosts.2, 3 While the adaptive immune response has been proved to be an essential component of cancer immunosurveillance,4, 5 accelerated tumor development in mice lacking natural killer (NK) cells or NK cell-activating receptors6, 7, 8, 9, 10, 11, 12 suggests that NK cells are also involved.
It has been reported that immunogenic antigens, such as Wilms tumor 113 or Proteinase 3,14 are expressed in human leukemia, and cytotoxic T cells recognizing these antigens were detected in leukemia patients.15 In addition, ligands for NK cell-activating receptors, such as NK group 2, member D (NKG2D) or DNAX accessory molecule 1 (DNAM-1), are expressed on human leukemia cells.16, 17, 18 Although these findings indicate that leukemia cells interact with immune systems during their development, it is still unclear how exactly the immune systems affect leukemia development, because only clinically evident leukemia cells, which may have been immunologically selected, are usually available for analysis in human leukemia or mouse leukemia models.
MLL/ENL is a fusion gene generated as a result of t (11;19) translocation,19 which is responsible for a part of human acute monocytic leukemia. The MLL/ENL mouse leukemia-initiating cells are established by transducing MLL/ENL into hematopoietic progenitor cells (HPCs).20, 21 As the MLL/ENL oncogene confers self-renewal potential to HPCs, clonal MLL/ENL-expressing HPCs (MLL/ENL-HPCs) can be established and expanded in vitro,21 thus making it possible to transfer clonal leukemia-initiating cells into multiple recipient mice. Although MLL/ENL-HPCs can induce acute monocytic leukemia upon transplantation into irradiated recipient mice, it is unclear whether irradiation is essential for the engraftment of leukemia cells. Irradiation of the recipient mice largely impairs the immune systems, and this makes it impossible to analyze immunosurveillance against leukemia development.
In this study, to clarify how the presence of immune cells affects leukemia development, MLL/ENL-HPCs were transferred into immune-competent or immune-deficient mice without any preconditioning including irradiation. We found that MLL/ENL leukemia-initiating cells could engraft and induce leukemia in non-irradiated recipients in the absence of both adaptive immune cells and NK cells, and that leukemia development was largely inhibited by NK cells even in the absence of adaptive immunity. These results suggest that NK cells perform crucial roles in immunosurveillance against leukemia.
Materials and methods
Balb/c mice (from 6- to 8-week-old, female) were purchased from CREA Japan (Osaka, Japan). Rag2−/− and Rag2−/−γc−/− mice were kind gifts from Irving L Weissman (Stanford University, Stanford, CA, USA). All animal experiments in this study were approved by the administrative panel on laboratory animal care in Osaka University.
Establishment of MLL/ENL-HPCs
MLL-ENL complementary DNA (cDNA)19 was subcloned into the pMSCV-IRES-GFP (pMSCV-internal ribosomal entry site-green fluorescent protein) vector. Retroviral stocks were produced from Plat-E packaging cell line22 by transient transfection using Lipofetamine 2000 (Invitrogen, Carlsbad, CA, USA). C-kit+ bone marrow (BM) cells from 6- to 8-week-old mice were enriched using anti-c-kit microbeads (Miltenyi Biotech, Bergisch Gradbach, Germany), and cultured for 16 h in RPMI 1640 supplemented with 10% fetal calf serum, 10 ng/ml stem cell factor, 10 ng/ml interleukin-3 (IL-3) and 10 ng/ml IL-6 (PeproTech, Rocky Hill, NJ, USA). Cells were then infected with retroviral supernatants in the presence of 4 mg/ml polybrene. Three days after infection, GFP+ cells were FACS sorted using FACS Aria II (BD Biosciences, San Jose, CA, USA), and seeded in the methylcellulose medium (M3231; Stem Cell Technologies, Vancouver, BC, Canada) containing granulocyte–macrophage colony-stimulating factor, stem cell factor, IL-3 and IL-6 (10 ng/ml for each). After 7 days of culture, colonies were pooled, and then 1 × 104 cells were replated. At the third round of culture, single colonies were plucked from methylcellulose and transferred to liquid culture in the media containing stem cell factor, IL-3 and IL-6 (10 ng/ml for each). The MLL/ENL-expressing immortalized HPCs, which we named MLL/ENL-HPCs, were expanded, and multiple frozen stocks were made to utilize in the following experiments. Three independent MLL/ENL-HPC clones were established and used for the analysis.
For detection of NKG2D ligands, cells were incubated with recombinant human NKG2D/CD314 Fc chimera (R&D Systems, Minneapolis, MN, USA), then with biotin-conjugated anti-human IgG (H&L; Rockland, Gilbertsville, PA, USA), and finally with PE conjugated streptavidin (BioLegend, San Diego, CA, USA). For other staining, the following antibodies were used: anti-mouse Qa-1[b]-biotin (δA8.δF10.1Aδ; BD Pharmingen, San Diego, CA, USA), anti-mouse CD49b-biotin (DX5; BioLegend), anti-mouse Rae-1α/β/γ (R&D Systems, Minneapolis, MN, USA), anti-mouse major histocompatibility complex (MHC) class I H2Ld (Acris Antibodies GmbH, Herford, Germany), anti-mNectin-2 (R&D Systems), anti-mouse MHC class I H2Kd-PE (SF1-1.1.1; eBioscience, San Diego, CA, USA), anti-mouse MHC class I H2Dd-PE (34-5-8S; Abcam, Cambridge, MA, USA), anti-mouse H60-PE (205326; R&D Systems), anti-M/R MULT1-PE (5D10; eBioscience), anti-mouse CD155-PE (TX56; BioLegend), anti-mouse CD48-PE/Cy7 (HM48-1; BioLegend), anti-mouse CD117-APC (2B8; BioLegend), anti-mouse CD11b-PE (M1/70; BioLegend) and anti-mouse CD3-Alexa 647 (17A2; BioLegend), streptavidin-PE, anti-rat IgG-PE (Poly4054; BioLegend). Analysis was performed on FACS Aria II or FACS Canto (BD Biosciences).
NK cell cytotoxicity assay
NK cells were isolated from BM and spleen of Balb/c mice using DX5 Microbeads (Miltenyi Biotech), and were cultured in RPMI 1640 supplemented with 10% fetal calf serum and IL-2 (65 IU/ml) for 7–10 days. Target cells were labeled with Chromium-51 (Cr51) by incubating for 90 min, and then cocultured with the activated NK cells for 4 h. An aliquot of supernatant was removed, and the amount of radioactivity was measured with a gamma counter. Percent of specific lysis was calculated as follows: ((experimental release−spontaneous release)/(maximum release−spontaneous release)) × 100. For the blocking experiments with anti-NKG2D antibody in vitro, anti-NKG2D mAb (CX-5; eBioscience) or IgG1 isotype control was added at the concentration of 10 μg/ml.
Transplantation of MLL/ENL-HPCs
In the transplantation with MLL/ENL-HPCs, 5 × 106 cells were retro-orbitally injected. In the secondary transplantation, 1 × 102−1 × 104 GFP+c-kit+Mac1+ BM cells from Rag2−/−γc−/− leukemia mice were transplanted. Peripheral blood (PB) cells were analyzed for the percentages of GFP+ leukemia cells on FACS Canto (Becton, Dickinson and Company, NYSE: BDX) every 2 weeks. Mice were diagnosed with leukemia when GFP+ leukemia cells accounted for more than 0.1% of mononuclear cells in PB.
In the engraftment analysis, mice were killed 24 h after transplantation and BM cells from both femurs and tibias or spleen cells were extracted and subjected to the flow cytometry analysis for counting GFP+PI− cells. More than 1 × 106 PI− cells were analyzed for each sample. Examination of 1 × 106 PI− BM cells from untreated wild-type mice did not find a single GFP-positive cell in the GFP+ fraction (Supplementary Figure 1). In the NK cell depletion experiments, 100 μl of anti-asialo GM1 polyclonal antibody (WAKO Pure Chemicals Industries Ltd, Osaka, Japan) were intravenously injected two times (days −1 and 0) before transplantation. In the NKG2D blocking experiments, 100 μg of anti-mouse NKG2D monoclonal antibody (CX5; eBioscience) or control rat IgG was intravenously injected two times (days −1 and 0) before transplantation. In the NK cell transfer experiment, Balb/c mice were intraperitonally administered with poly I:C (10 μg), and the next day, NK cells were purified from splenocytes using DX5 microbeads. Purified NK cells were cultured in RPMI 1640 supplemented with 10% fetal calf serum and IL-2 (65 IU/ml) for 7–10 days, and then transferred into the mice by retro-orbital injection 1 day before or 2–4 days after transplantation of MLL/ENL-HPCs.
RNA extraction, cDNA synthesis and quantitative RT-PCR
Total RNAs were isolated using the TRIZOL (Invitrogen). cDNAs were synthesized using SuperScript III (Life Technologies, Carlsbad, CA, USA). Quantitative reverse transcription-PCR (qRT-PCR) was performed using Power SYBER Green PCR master mix (Applied Biosystems, Foster, CA, USA) on Applied Biosystems 7900HT Fast Real-time PCR system. Data analyses were performed using the delta-delta Ct methods.
Primer sequences used for each gene were as follows: MLL, 5′-IndexTermGCAAACAGAAAAAAGTGGCTCCCCG-3′; ENL, IndexTerm5′-TCATGTGGCCACGGCCTCCAG-3′; Rae-1α, 5′-IndexTermGGAAGTTGATGAATTCTTAAAGCAG-3′, 5′-IndexTermGAAGTAGAGTGGCTGGGAGGT-3′; Rae1-β, 5′-IndexTermTTGATACACGAATGCAGTCAGA-3′, 5′-IndexTermGAAGCGGGGAAGTTGATGTA-3′; Rae1-γ, 5′-IndexTermCATACCCGAATGCAGACAGA-3′, 5′-IndexTermGAGGTGGAAGTGGGGAAGTT-3′; H60, 5′-IndexTermTGTGGAAGTTACAATCAGGTAATGA-3′, 5′-IndexTermGATGCTATGCTGTTCATCAGTGT-3′; CD112, 5′-IndexTermGATGTACGAAGCAACCCAGAG-3′, 5′-IndexTermTCCACAGAGTGGACAAGCAG-3′; CD155, 5′-IndexTermTACGAAGATGCAGGCATGG-3′, 5′-IndexTermCTGTTTGGCTCCATGTTCAG-3′; CD48, 5′-IndexTermTTACACCTGCCAAGTCAGCA-3′, 5′-IndexTermACCACTAGCCAAGTTGCAGTC-3′; β-actin, 5′-IndexTermTGTCCACCTTCCAGCAGATGT-3′, 5′-IndexTermAGCTCAGTAACAGTCCGCCTAGA-3′.
Two-sample t-test was used for the statistical analysis in all experiments. Differences were defined as statistically significant when P-values were <0.05.
MLL/ENL-HPCs are rejected in non-irradiated wild-type mice
MLL/ENL leukemia-initiating cells were generated as reported previously.23, 24 MLL/ENL-transduced c-kit+ BM cells were cultured in semisolid medium supplemented with stem cell factor, IL-3 and IL-6, followed by serially replating of the resultant colonies. After the third round of replating, MLL/ENL-HPCs were expanded by isolating a single colony and culturing it in liquid media containing the cytokines (Figure 1a). MLL/ENL-HPCs were immature myeloid cells, most of which were positive for c-kit and Mac-1 (Figures 1b and c). When MLL/ENL-HPCs were transplanted into lethally (9 Gy) irradiated wild-type mice, all recipients were diagnosed with leukemia (>0.1% of GFP+ cells in PB) within 1 month (Figure 1d), and died from 80 days to 110 days after transplantation (Figure 1e). Immature and mature myeloid cells, which were considered likely to correspond to c-kit+Mac-1+ and c-kit+Mac-1− cells, greatly increased in BM of the leukemia mice (Figures 1f and g). In contrast, when MLL/ENL-HPCs were transplanted into non-irradiated mice, none of the recipients developed leukemia for 140 days (Figures 1d and e). Moreover, no GFP+ leukemia cells were detected in the BM and spleen on day 140. These results show that MLL/ENL-HPCs can engraft and induce leukemia in lethally irradiated wild-type mice, but are rejected in non-irradiated ones.
NK cells inhibit MLL/ENL leukemia development even in the absence of adaptive immunity
To determine how the immune cells were involved in the rejection of leukemia cells, 5 × 106 MLL/ENL-HPCs were intravenously transferred into the wild-type, Rag2−/− (lacking T, B and NKT cells) or Rag2−/−γc−/− (lacking T, B, NKT and NK cells) recipient mice without any preconditioning, and PB cells were analyzed every 2 weeks (Figure 2a). On day 28, no GFP+ leukemia cells were detected in any of the wild-type and Rag2−/− recipients, but all of the Rag2−/−γc−/− recipients were diagnosed with leukemia (Figure 2b). No leukemia development had been detected in any of the wild-type recipient and three out of five Rag2−/− recipients by day 140, whereas two Rag2−/− recipients had developed leukemia on day 112 (Figure 2c). All disease-free mice were killed on day 140, and no GFP+ cells were detected in their BM. Leukemia development was always confirmed by May–Grünwald–Giemsa staining of BM specimens, FACS analysis and PCR detection of the MLL/ENL transgene in BM cells (Figures 2d–f). No differences were observed among the characteristics of leukemia cells developed in irradiated wild-type, Rag2−/− and Rag2−/−γc−/− recipients.
It has been reported that far fewer MLL/ENL leukemia cells were needed to be transplanted to induce leukemia in the secondary transplant setting.24 Thus, 1 × 102, 1 × 103 or 1 × 104 GFP+ MLL/ENL leukemia cells that had developed in Rag2−/−γc−/− recipients were used for secondary transplantation into the wild-type, Rag2−/− or Rag2−/−γc−/− recipient mice without any preconditioning (Figure 2g). As few as 1 × 102 MLL/ENL leukemia cells could induce leukemia in the Rag2−/−γc−/− recipient mice within 4 weeks (Figure 2h). However, 1 × 102 and 1 × 103 leukemia cells were rejected in all the Rag2−/− and wild-type recipients (Figure 2i). When 1 × 104 leukemia cells were transplanted, all the wild-type recipients still rejected leukemia cells, whereas two of three Rag2−/− recipients developed leukemia after longer latency periods than needed for the Rag2−/−γc−/− recipients (Figure 2i).
These results clearly indicated that the immune system contributed to the rejection of MLL/ENL leukemia cells in the wild-type recipient mice. Furthermore, these results also suggested that NK cells played a crucial role for the rejection of leukemia.
NK cells can eliminate MLL/ENL leukemia cells in vivo in the absence of adaptive immunity
NK cell-mediated elimination of transplanted cells was reported to occur within 24 h,25 which led us to examine whether transplanted MLL/ENL-HPCs were also eliminated within 24 h after transplantation. Rag2−/− or Rag2−/−γc−/− mice were transplanted with 5 × 106 MLL/ENL-HPCs, and GFP+ leukemia cells in BM or spleen were counted by means of flow cytometry 24 h after transplantation (Figure 3a). In BM, 0.4 and 18.6 GFP+ leukemia cells per 1 × 106 cells were detected in Rag2−/− and Rag2−/−γc−/− recipients, respectively (Figures 3b and c, P<0.05). The corresponding values for the spleen were 0.3 and 152.2 GFP+ leukemia cells (Figure 3c, P<0.05). These results suggest that most MLL/ENL-HPCs were eliminated within 24 h after transplantation in the presence of NK cells without adaptive immunity.
We next examined whether NK cell depletion from Rag2−/− mice enhanced the survival of MLL/ENL-HPCs in vivo. To this end, Rag2−/− mice were pretreated with anti-asialo GM1 antibody (100 μg per day, days −1 and 0) to deplete NK cells, and then transplanted with 5 × 106 MLL/ENL-HPCs (Figures 3d and e). GFP+ leukemia cells in BM were counted by means of flow cytometry 24 h after transplantation. Significantly more GFP+ leukemia cells were found in the mice treated with anti-asialo GM1 antibody than in the untreated ones (Figure 3f, 20.6 vs 0.3 per 1 × 106 cells, P<0.05), whereas the number of engrafted leukemia cells in the Rag2−/− mice treated with anti-asialo GM1 antibody were comparable to those in the Rag2−/−γc−/− recipients (Figure 3f, 20.6 vs 18.6 per 1 × 106 cells, P=0.42). Furthermore, we performed the same experiments in the secondary transplant setting with smaller numbers of transplanted cells. MLL/ENL leukemia cells (5 × 105 cells) from Rag2−/−γc−/− recipient mice had been transplanted into secondary Rag2−/− recipients with or without anti-asialo GM1 antibody pretreatment. Significantly more GFP+ leukemia cells were found in mice treated with anti-asialo GM1 antibody than in untreated ones (Figure 3g, 15.0 vs 0.2 per 1 × 106 cells, P<0.05). Taken together, these results indicated that the elimination of leukemia cells in the Rag2−/− recipient mice mostly depend on the presence of NK cells.
We also tested whether leukemia cells could be eliminated by transfer of NK cells into Rag2−/−γc−/− recipient mice (Figure 3h). Dx5-positive NK cells were purified from the wild-type mice and were then cultured in IL2-containing medium for 7–10 days. The percentages of TCR−Dx-5+CD122+ NK cells26 in the NK cell preparations were more than 90% after in vitro culture (Figure 3i). In the first experiment, 1 × 106 NK cells were transferred into Rag2−/−γc−/− mice, followed the next day by transplantation of 5 × 106 MLL/ENL-HPCs. GFP+ leukemia cells in BM were counted 24 h later (Figure 3h, Exp. 1). As shown in Figure 3j, significantly fewer GFP+ leukemia cells were detected in the mice with transferred NK cells than in the untreated ones (4.7 vs 18.7 per 1 × 106 cells, P<0.05). Next, to examine whether NK cells could eliminate leukemia cells that had already engrafted, 1 × 106 MLL/ENL-HPCs were transplanted into Rag2−/−γc−/− mice, followed by the transfer of 2 × 106 NK cells on day 2–4. GFP+ leukemia cells in BM were counted on day 7 (Figure 3h, Exp. 2). As shown in Figure 3k, significantly fewer GFP+ leukemia cells were detected in the mice with NK cell transfer than in untreated ones (9.2 vs 87.2 per 1 × 106 cells, P<0.05). Taken together, these results constitute clear evidence that NK cells can eliminate MLL/ENL-HPCs in vivo even in the absence of adaptive immunity.
NK cells recognize NKG2D ligands on MLL/ENL-HPCs
Next, we used the Cr51 release assay to determine whether MLL/ENL-HPCs could be recognized and killed by NK cells in vitro. MLL/ENL-HPCs or freshly purified normal HPCs were used as target cells and cocultured for 4 h with NK cells stimulated with IL2. MLL/ENL-HPCs proved to be more sensitive to cytotoxicity by NK cells than normal HPCs (Figure 4a).
The activation of NK cells is regulated by the balance between inhibitory signals, transduced mainly by the interaction of MHC class I molecules and killer inhibitory receptors receptors, and activating signals, mediated by activating NK cell receptors.27 Expressions of MHC class I molecules (H2Kd, H2Dd, H2Ld and Qa-1) on MLL/ENL-HPCs and normal HPCs were compared, but no significant differences were found (Figure 4b). It was reported that NKG2D, DNAM1 and 2B4 receptors have important roles in tumor cell recognition.28, 29, 30 We used qRT-PCR to measure the expression of NKG2D ligands (Rae-1α/β/γ, H60 and MULT1), DNAM-1 ligands (CD112 and CD155) and a 2B4 ligand (CD48) on MLL/ENL-HPCs and normal HPCs (Figure 4c). Expression levels of Rae-1α, Rae-1β, Rae-1γ, H60 and MULT1 in MLL/ENL-HPCs were 274, 128, 137, 8.6 and 1.6 times higher, respectively, than those in normal HPCs. Expression levels of CD112 and CD155 in MLL/ENL-HPCs were also 6.5 and 12.1 times higher compared with those in normal HPCs. In contrast, expression of CD48, which is reportedly an inhibitory ligand of 2B4,31 was 9.2 times lower in MLL/ENL-HPCs. Flow cytometry analysis of the ligands for NK cell receptors showed that higher expressions of NKG2D/DNAM1 ligands and reduced expression of CD48 were also found on all the MLL/ENL-HPC clones at the protein level (Figure 4d).
To clarify whether NK cells recognized NKG2D ligands on MLL/ENL leukemia cells and killed them, the effect of NKG2D blockade on the elimination of leukemia cells by NK cells was examined. An in vitro NK cell cytotoxicity assay indicated that lysed leukemia cells were significantly reduced by the addition of anti-NKG2D antibody (Figure 4e). The effect of NKG2D blocking was also tested in an in vivo engraftment assay. Rag2−/− mice treated with anti-NKG2D antibody or control rat IgG (100 μg, days −1 and 0) were transplanted with 5 × 106 MLL/ENL-HPCs (Figure 4f). In BM of the mice treated with anti-NKG2D antibody and those with control rat IgG, 4.9 and 0.3 GFP+ leukemia cells per 1 × 106 cells, respectively, were detected 24 h after transplantation (Figures 4g, P<0.05). These results indicate that NKG2D is involved in the recognition and elimination of leukemia cells by NK cells not only in vitro but also in vivo.
MLL/ENL induces the expression of NK cell-activating ligands
Next, we examined whether the expression of NK cell-activating ligands was induced by MLL/ENL expression. Either MSCV-IRES-GFP (Mig, empty vector) or MSCV-MLL/ENL-IRES-GFP (Mig-MLL/ENL) was transduced into c-kit+ BM cells, and FACS-sorted GFP+ cells were cultured in methylcellulose media supplemented with cytokines. Resultant colonies were collected and analyzed for the expression of ligands for NK cell receptors (Figure 5a). Expression of NKG2D ligands, especially Rae-1α/β/γ, was significantly higher in MLL/ENL-transduced than in empty vector-transduced HPCs. DNAM1 ligand expression was also found to be significantly higher in MLL/ENL-transduced than in empty vector-transduced HPCs, whereas CD48 expression was lower in MLL/ENL-transduced HPCs (Figures 5b and c). These findings indicate that upregulation of NKG2D/DNAM-1 ligands and downregulation of CD48 were caused by MLL/ENL expression, rather than by retroviral transduction or in vitro culture procedure.
MLL/ENL leukemia cells developed in the presence of NK cells are resistant to NK cell-mediated elimination
To examine whether leukemia cells resistant to NK cell recognition were selected during leukemia development, the expression levels of the ligands for NK cell receptors were compared between leukemia cells that had developed in the Rag2−/− recipients and those in the Rag2−/−γc−/− recipients (Figure 6a). Results of the qRT-PCR analysis showed that leukemia cells developed in the Rag2−/− recipients expressed lower levels of NKG2D ligands and DNAM-1 ligands and higher levels of CD48 than did those developed in the Rag2−/−γc−/− recipients (Figure 6b). Flow cytometry analysis further showed that NKG2D ligands, especially Rae1α/β/γ, were downregulated on leukemia cells developed in the Rag2−/− recipients (Figure 6c). Leukemia cells developed in the Rag2−/− or Rag2−/−γc−/− recipients were then secondary transplanted into unconditioned Rag2−/− recipients. Consistent with the expressions of the ligands for NK cell receptors, larger numbers of leukemia cells obtained from Rag2−/− recipients could survive and engraft in BM 24 h after transplantation than could those obtained from Rag2−/−γc−/− recipients (Figure 6d, 8.4 vs 0.2 per 1 × 106 cells, P<0.05). In addition, latency periods for leukemia development were shorter for the mice transplanted with leukemia cells from Rag2−/− recipients than for those transplanted with leukemia cells from Rag2−/−γc−/− recipients (Figure 6e). These results suggest that leukemia cells expressing low levels of ligands for NKG2D and DNAM-1 selectively survived during leukemia development in Rag2−/− recipients.
In this study, we found that MLL/ENL-HPCs could engraft and induce leukemia in non-irradiated Rag2−/−γc−/− recipients that lacked both adaptive immune cells and NK cells, but not in wild-type recipients. As MLL/ENL-HPCs express the human MLL/ENL gene and GFP, both of which are xenoantigens for mice, we thought at first that the rejection was mediated by xenoantigen-specific T-cell reaction as previously reported for transplantation with GFP-marked cells.32, 33, 34 However, our analysis performed in this study revealed that NK cells are crucial for the rejection of MLL/ENL leukemia cells.
We were able to show that NK cells could eradicate MLL/ENL leukemia cells in vivo in the absence of adaptive immunity. It has been reported that NK cell-activating ligands is intrinsic sensors of oncogenic transformation and induce a signal for innate immune surveillance.35 In addition, it was demonstrated that NK cells eliminated syngeneic cells with enforced expression of an NK cell-activating ligand in vivo.36, 37, 38 These findings suggest that NK cells have the potential to recognize and eliminate developing tumor cells at early stages of tumor development. However, it has been unclear whether NK cells directly kill tumor cells or eliminate tumor cells by secreting proinflammatory cytokines such as IFN-γ, TNF-α and chemokines, and thus activating an adaptive immune response.39, 40 In most of the previous studies, the function of NK cells has been examined in the presence of adaptive immune cells. However, the results of our study clearly demonstrated that oncogene-driven leukemia cells can be eliminated by NK cells in vivo even in the absence of adaptive immune cells. It thus appears that NK cells function as the first line of defense against leukemia development by eliminating nascent leukemia cells before they increase enough to stimulate adaptive immune responses. Our results support the notion that NK cell-activating therapies41, 42 may constitute an attractive option for prevention of leukemia development.
Activating ligands for NKG2D and DNAM-1 were upregulated and CD48 was downregulated in MLL/ENL-HPCs compared with that in control vector-transduced HPCs. This suggests that the expression of the MLL/ENL oncogene can transform non-immunogenic hematopoietic cells into immunogenic ones, and stimulate NK cell-mediated immunosurveillance. NK cells may thus eliminate leukemia cells at the very early stages of leukemia development to prevent further development. However, our results also suggest that leukemia cells expressing low levels of ligands for NKG2D and DNAM-1 selectively survived in the presence of NK cells, which implies that the NK cell-mediated immunoediting occurs. It was recently reported that certain drugs, such as histone deacetylase inhibitors,43 demethylating agents44 and proteasome inhibitors,45 upregulated NKG2D ligands or DNAM-1 ligands. Our results thus support the notion that upregulation of NK cell-activating ligands by these immunomodulatory drugs may constitute a therapeutic strategy for the prevention of leukemia development.
Transplant-based mouse leukemia models are widely used in leukemia studies. However, many researchers have not paid much attention to immune reactions to transplanted leukemia cells. In our study, we used non-irradiated mice as recipients and showed that immune response played a substantial role in the rejection of transplanted leukemia cells. The MLL/ENL mouse leukemia model for non-irradiated recipients thus offers a viable basis for the analysis of immune response to leukemia cells. Different from solid tumor models, leukemia cells can be easily sampled for analysis at any time points during leukemia development, thus making it possible to analyze the characteristics of both immune cells and leukemia cells over time. It should also be noted that MLL/ENL-HPCs are for the most part different from tumor cell lines in that they possess the potential not only to initiate leukemia but also differentiate into mature progenies in vivo. In addition, as MLL/ENL-HPCs can self-renew in vitro, unlimited numbers of clonal leukemia-initiating cells can be generated for transplantation into multiple recipient mice. This mouse leukemia model may thus be useful not only for analysis of NK cell-mediated immunosurveillance but also of T cell-mediated immune response as well as for evaluation of many types of immunomodulatory therapies.
In summary, the results reported here represent the first demonstration of the potential of NK cells to eradicate oncogene-driven leukemia cells in vivo even in the absence of adaptive immune responses. Our findings offer support for the possibility that NK cells function as the first line of defense against leukemia development, and suggest that therapies involving activation of NK cells may be an option for prevention of development or relapse of leukemia.
Burnet M . Cancer: a biological approach III. Viruses associated with neoplastic conditions. IV. Practical applications. BMJ 1957; 1: 841–847.
Dunn GP, Bruce AT, Ikeda H, Old LJ, Schreiber RD . Cancer immunoediting: from immunosurveillance to tumor escape. Nat Immunol 2002; 3: 991–998.
Schreiber RD, Old LJ, Smyth MJ . Cancer immunoediting: integrating immunity's roles in cancer suppression and promotion. Science 2011; 331: 1565–1570.
Matsushita H, Vesely MD, Koboldt DC, Rickert CG, Uppaluri R, Magrini VJ et al. Cancer exome analysis reveals a T-cell-dependent mechanism of cancer immunoediting. Nature 2012; 482: 400–404.
DuPage M, Mazumdar C, Schmidt LM, Cheung AF, Jacks T . Expression of tumour-specific antigens underlies cancer immunoediting. Nature 2012; 482: 405–409.
van den Broek ME, Kagi D, Ossendorp F, Toes R, Vamvakas S, Lutz WK et al. Decreased tumor surveillance in perforin-deficient mice. J Exp Med 1996; 184: 1781–1790.
Street SE, Cretney E, Smyth MJ . Perforin and interferon-gamma activities independently control tumor initiation, growth, and metastasis. Blood 2001; 97: 192–197.
Cretney E, Takeda K, Yagita H, Glaccum M, Peschon JJ, Smyth MJ . Increased susceptibility to tumor initiation and metastasis in TNF-related apoptosis-inducing ligand-deficient mice. J Immunol 2002; 168: 1356–1361.
Guerra N, Tan YX, Joncker NT, Choy A, Gallardo F, Xiong N et al. NKG2D-deficient mice are defective in tumor surveillance in models of spontaneous malignancy. Immunity 2008; 28: 571–580.
Gilfillan S, Chan CJ, Cella M, Haynes NM, Rapaport AS, Boles KS et al. DNAM-1 promotes activation of cytotoxic lymphocytes by nonprofessional antigen-presenting cells and tumors. J Exp Med 2008; 205: 2965–2973.
Iguchi-Manaka A, Kai H, Yamashita Y, Shibata K, Tahara-Hanaoka S, Honda S et al. Accelerated tumor growth in mice deficient in DNAM-1 receptor. J Exp Med 2008; 205: 2959–2964.
Elboim M, Gazit R, Gur C, Ghadially H, Betser-Cohen G, Mandelboim O . Tumor immunoediting by NKp46. J Immunol 2010; 184: 5637–5644.
Oka Y, Tsuboi A, Taguchi T, Osaki T, Kyo T, Nakajima H et al. Induction of WT1 (Wilms' tumor gene)-specific cytotoxic T lymphocytes by WT1 peptide vaccine and the resultant cancer regression. Proc Natl Acad Sci USA 2004; 101: 13885–13890.
Molldrem JJ, Clave E, Jiang YZ, Mavroudis D, Raptis A, Hensel N et al. Cytotoxic T lymphocytes specific for a nonpolymorphic proteinase 3 peptide preferentially inhibit chronic myeloid leukemia colony-forming units. Blood 1997; 90: 2529–2534.
Molldrem JJ, Lee PP, Wang C, Champlin RE, Davis MM . A PR1-human leukocyte antigen-A2 tetramer can be used to isolate low-frequency cytotoxic T lymphocytes from healthy donors that selectively lyse chronic myelogenous leukemia. Cancer Res 1999; 59: 2675–2681.
Salih HR, Antropius H, Gieseke F, Lutz SZ, Kanz L, Rammensee HG et al. Functional expression and release of ligands for the activating immunoreceptor NKG2D in leukemia. Blood 2003; 102: 1389–1396.
Pende D, Spaggiari GM, Marcenaro S, Martini S, Rivera P, Capobianco A et al. Analysis of the receptor-ligand interactions in the natural killer-mediated lysis of freshly isolated myeloid or lymphoblastic leukemias: evidence for the involvement of the Poliovirus receptor (CD155) and Nectin-2 (CD112). Blood 2005; 105: 2066–2073.
Hilpert J, Grosse-Hovest L, Grunebach F, Buechele C, Nuebling T, Raum T et al. Comprehensive analysis of NKG2D ligand expression and release in leukemia: implications for NKG2D-mediated NK cell responses. J Immunol 2012; 189: 1360–1371.
Tkachuk DC, Kohler S, Cleary ML . Involvement of a homolog of Drosophila trithorax by 11q23 chromosomal translocations in acute leukemias. Cell 1992; 71: 691–700.
Slany RK, Lavau C, Cleary ML . The oncogenic capacity of HRX-ENL requires the transcriptional transactivation activity of ENL and the DNA binding motifs of HRX. Mol Cell Biol 1998; 18: 122–129.
Cozzio A, Passegue E, Ayton PM, Karsunky H, Cleary ML, Weissman IL . Similar MLL-associated leukemias arising from self-renewing stem cells and short-lived myeloid progenitors. Genes Dev 2003; 17: 3029–3035.
Morita S, Kojima T, Kitamura T . Plat-E: an efficient and stable system for transient packaging of retroviruses. Gene Ther 2000; 7: 1063–1066.
Lavau C, Szilvassy SJ, Slany R, Cleary ML . Immortalization and leukemic transformation of a myelomonocytic precursor by retrovirally transduced HRX-ENL. EMBO J 1997; 16: 4226–4237.
Somervaille TC, Cleary ML . Identification and characterization of leukemia stem cells in murine MLL-AF9 acute myeloid leukemia. Cancer Cell 2006; 10: 257–268.
Hamby K, Trexler A, Pearson TC, Larsen CP, Rigby MR, Kean LS . NK cells rapidly reject allogeneic bone marrow in the spleen through a perforin- and Ly49D-dependent, but NKG2D-independent mechanism. Am J Transplant 2007; 7: 1884–1896.
Di Santo JP . Natural killer cell developmental pathways: a question of balance. Annu Rev Immunol 2006; 24: 257–286.
Vivier E, Ugolini S, Blaise D, Chabannon C, Brossay L . Targeting natural killer cells and natural killer T cells in cancer. Nat Rev Immunol 2012; 12: 239–252.
Raulet DH . Roles of the NKG2D immunoreceptor and its ligands. Nat Rev Immunol 2003; 3: 781–790.
Raulet DH, Guerra N . Oncogenic stress sensed by the immune system: role of natural killer cell receptors. Nat Rev Immunol 2009; 9: 568–580.
Vivier E, Raulet DH, Moretta A, Caligiuri MA, Zitvogel L, Lanier LL et al. Innate or adaptive immunity? The example of natural killer cells. Science 2011; 331: 44–49.
Mooney JM, Klem J, Wulfing C, Mijares LA, Schwartzberg PL, Bennett M et al. The murine NK receptor 2B4 (CD244) exhibits inhibitory function independent of signaling lymphocytic activation molecule-associated protein expression. J Immunol 2004; 173: 3953–3961.
Stripecke R, Carmen Villacres M, Skelton D, Satake N, Halene S, Kohn D . Immune response to green fluorescent protein: implications for gene therapy. Gene Ther 1999; 6: 1305–1312.
Rosenzweig M, Connole M, Glickman R, Yue SP, Noren B, DeMaria M et al. Induction of cytotoxic T lymphocyte and antibody responses to enhanced green fluorescent protein following transplantation of transduced CD34(+) hematopoietic cells. Blood 2001; 97: 1951–1959.
Follenzi A, Battaglia M, Lombardo A, Annoni A, Roncarolo MG, Naldini L . Targeting lentiviral vector expression to hepatocytes limits transgene-specific immune response and establishes long-term expression of human antihemophilic factor IX in mice. Blood 2004; 103: 3700–3709.
Unni AM, Bondar T, Medzhitov R . Intrinsic sensor of oncogenic transformation induces a signal for innate immunosurveillance. Proc Natl Acad Sci USA 2008; 105: 1686–1691.
Cerwenka A, Baron JL, Lanier LL . Ectopic expression of retinoic acid early inducible-1 gene (RAE-1) permits natural killer cell-mediated rejection of a MHC class I-bearing tumor in vivo. Proc Natl Acad Sci USA 2001; 98: 11521–11526.
Diefenbach A, Jensen ER, Jamieson AM, Raulet DH . Rae1 and H60 ligands of the NKG2D receptor stimulate tumour immunity. Nature 2001; 413: 165–171.
Ogasawara K, Benjamin J, Takaki R, Phillips JH, Lanier LL . Function of NKG2D in natural killer cell-mediated rejection of mouse bone marrow grafts. Nat Immunol 2005; 6: 938–945.
Smyth MJ, Hayakawa Y, Takeda K, Yagita H . New aspects of natural-killer-cell surveillance and therapy of cancer. Nat Rev Cancer 2002; 2: 850–861.
Zhang C, Zhang J, Tian Z . The regulatory effect of natural killer cells: do ‘NK-reg cells’ exist? Cell Mol Immunol 2006; 3: 241–254.
Benson DM Jr., Bakan CE, Zhang S, Collins SM, Liang J, Srivastava S et al. IPH2101, a novel anti-inhibitory KIR antibody, and lenalidomide combine to enhance the natural killer cell versus multiple myeloma effect. Blood 2011; 118: 6387–6391.
Vey N, Bourhis JH, Boissel N, Bordessoule D, Prebet T, Charbonnier A et al. A phase 1 trial of the anti-inhibitory KIR mAb IPH2101 for AML in complete remission. Blood 2012; 120: 4317–4323.
Diermayr S, Himmelreich H, Durovic B, Mathys-Schneeberger A, Siegler U, Langenkamp U et al. NKG2D ligand expression in AML increases in response to HDAC inhibitor valproic acid and contributes to allorecognition by NK-cell lines with single KIR-HLA class I specificities. Blood 2008; 111: 1428–1436.
Tang KF, He CX, Zeng GL, Wu J, Song GB, Shi YS et al. Induction of MHC class I-related chain B (MICB) by 5-aza-2'-deoxycytidine. Biochem Biophys Res Commun 2008; 370: 578–583.
Soriani A, Zingoni A, Cerboni C, Iannitto ML, Ricciardi MR, Di Gialleonardo V et al. ATM-ATR-dependent up-regulation of DNAM-1 and NKG2D ligands on multiple myeloma cells by therapeutic agents results in enhanced NK-cell susceptibility and is associated with a senescent phenotype. Blood 2009; 113: 3503–3511.
We thanks Irving L Weissman (Stanford University) for kind gifts of Rag2−/−γc−/− mice, Michael L Cleary (Stanford University) for kind gift of MLL/ENL cDNA, Toshio Kitamura (Tokyo University) for kind gift of Plat-E cells, Hisashi Arase (Osaka University) for fruitful discussion and Ruriko Inoue (Osaka University) for technical assistance.
The authors declare no conflict of interest.
Supplementary Information accompanies this paper on the Leukemia website
About this article
Cite this article
Nakata, J., Nakano, K., Okumura, A. et al. In vivo eradication of MLL/ENL leukemia cells by NK cells in the absence of adaptive immunity. Leukemia 28, 1316–1325 (2014) doi:10.1038/leu.2013.374
- acute myeloid leukemia
- NK cell
- tumor immunity
Cellular and Molecular Life Sciences (2018)
Life Science Alliance (2018)
Critical Reviews in Oncology/Hematology (2016)
NK cells play a significant role in immunosurveillance at the early stage of MLL-AF9 acute myeloid leukemia via CD226/CD155 interactions
Science China Life Sciences (2015)
An Immunocompetent Mouse Model for MLL/AF9 Leukemia Reveals the Potential of Spontaneous Cytotoxic T-Cell Response to an Antigen Expressed in Leukemia Cells
PLOS ONE (2015)