Natural killer (NK) cell adoptive transfer is a promising approach for cancer immunotherapy; however, its development has been hindered by the lack of efficient methods to produce large numbers of functional NK cells. In this study, we engineered the leukaemia cell line K562 to express CD137 ligand (CD137L) and membrane-bound interleukin (mbIL)-21 on the cell surface, and used these cells to expand NK cells from the peripheral blood mononuclear cells. We found that purity of the NK cells (CD3−CD56+/CD16+) increased from less than 30% to above 95% after a 3-week expansion and proliferation of the cells was sustained for more than 8 weeks. The surface expression of NK cell activating and inhibitory receptors, except for NKp80, was clearly increased with the expansion, and NK cell-mediated killing activity was also enhanced significantly. However, these changes in both phenotype and function were clearly reversed by JSI-124, a specific signal transducer and activator of transcription-3 (STAT-3) inhibitor. Taken together, data showed that the combination of mbIL-21 and CD137L could efficiently induce the formation of functional human NK cells from peripheral blood mononuclear cells, and STAT-3 inhibition could impair this induction. Therefore, STAT-3 activation may benefit human NK cell proliferation and cytotoxicity, and provide valuable clinical applications in NK cell immunotherapy against viral infectious diseases and cancers.
Human natural killer (NK) cells are a subset of peripheral blood lymphocytes that are defined by their expression of CD56 and/or CD16 and the absence of T cell receptor CD3 . NK cells can recognize and subsequently kill virus-infected and transformed cells in the absence of prior stimulation, and play a critical role in the immune surveillance of virus infectious diseases and cancers. NK cell killing is regulated through balanced signals from the activating and inhibitory receptors on NK cell surface . A large number of studies have demonstrated that NK cells could elicit strong anti-tumour efficacy, and are promising effectors for adoptive immunotherapy against cancers .
NK cell alloreactivity could control leukaemia relapse without causing graft-versus-host disease (GVHD) . Adoptive transfer of NK cells has been tested in early-phase clinical trials and has emerged as a safe and potentially efficacious immunotherapy for cancers . However, development of adoptive NK cell immunotherapy has been hampered by insufficient numbers of NK cells available from donors, as NK cells represent only a small fraction of the peripheral blood mononuclear cells which are already in low numbers. Hence, NK cell-based therapies would benefit greatly from reliable methods that can produce large numbers of functional NK cells ex vivo.
Several groups have demonstrated that the combination of activating signals provided by the K562 cell line, co-stimulation via 4-1BBL (CD137L) and survival signals provided by cytokines can mediate NK cell proliferation, such as the expansion of highly cytotoxic human NK cells, has been developed by modification of an artificial antigen-presenting cell line to induce expression of a membrane-bound form of interleukin (IL)-15 (mIL-15) and CD137 ligand . In this study, we directly modified K562 to express a membrane-bound form of IL-21 (mbIL-21) and CD137 ligand (CD137L). We found that the combination of mbIL-21-CD137L-K562 cells induced high-purity functional NK cells with sustained proliferation and high cytotoxicity from peripheral blood mononuclear cells through specific signal transducer and activator of transcription-3 (STAT-3) activation. Our results demonstrated the effectiveness of this simple method to generate large numbers of functional human NK cells, and elucidated that STAT-3 activation is required for human NK cell proliferation and cytotoxicity.
Methods and materials
The IL-21-Fc(CoOP)-pSBSO plasmid containing human Fc and membrane-bound regions, and the GlySer-EGFP(CoOp)-pSBSO sleeping beauty transposon expression vector, were gifted from Dr Laurence J. N. Cooper at the University of Texas MD Anderson Cancer Center. The CD137L/PCR4 TOPO® vector was purchased from Open Biosysems (Thermo Fisher Scientific, Inc., Waltham, MA, USA). The CD137L/pSBSO sleeping beauty expression vector was constructed by inserting the polymerase chain reaction (PCR) fragment derived from CD137L/PCR4 TOPO into the Nhe I-Xho I cloning site of the GlySer-EGFP(CoOp)-pSBSO vector. The forward primer of CD137L was 5′-AATGCTAGCGCCACCATGGAATACGCCTCTGACGC-3′; and the reverse primer was 5′-AAACTCGAGTTATTCCGACCTCGGTGAAGG-3′. The SB11 transponsase was obtained from the University of Texas MD Anderson Cancer Center via a material transfer agreement.
The antibodies [phycoerythrin (PE) anti-human CD137L, PE anti-human IL-21, allophycocyanin (APC) anti-human CD56, fluorescein isothiocyanate (FITC) anti-human CD3, PE anti-human CD16, PE anti-human NKG2D, PE anti-human NKp30, PE anti-human NKp44, PE anti-human NKp46, PE anti-human NKp80, PE anti-human CD226, PE anti-human 2B4, FITC anti-human KIR2DL1, FITC anti-human KIR2DL2 and FITC anti-human KIR3DL1], murine isotype controls [immunoglobulin [(Ig)G1κ-PE, IgG1κ-FITC, IgG2a –APC] and 7-amino-actinomycin D (7-AAD) were purchased from BioLegend, Inc. (San Diego, CA, USA). The recombinant human IL-2 protein was obtained from PeproTech (Rehovot, Israel). Calcein-acetoxymethylester (AM) was purchased from Sigma-Aldrich (St Louis, MO, USA).
Genetic engineering of K562 cells
K562 cells from ATCC were cultured in RPMI-1640 medium (Gibco, Carlsbad, CA, USA) supplemented with 10% fetal calf serum (Gibco), 1% penicillin–streptomycin and 2 mM of L-glutamine in 5% CO2 at 37°C. CD137L/pSBSO and SB11 were co-transfected into K562 cells using Lipofectmin 2000 (Invitrogen, Carlsbad, CA, USA), according to the manufacturer's instructions. The transfected K562 cells were cultured for 3 weeks, and then stained with FITC anti-human CD137L antibody. CD137L-positive K562 cells (CD137L-K562) were sorted by the fluorescence activated cell sorter (FACS)array II cytometer (BD Biosciences, San Jose, CA, USA) and continued to culture for another 2 weeks, then sorted again. After that, IL-21-Fc(CoOP)-pSBSO was transfected into CD137L-K562 cells together with SB11. Transfected CD137L-K562 cells were cultured for 3 weeks, and then stained with PE anti-human IL-21 antibody. IL-21-positive CD137L-K562 cells (mbIL-21-CD137L-K562) were sorted by the FACSarray II cytometer and continued to culture for another 2 weeks before sorted again.
NK cell expansion
Human peripheral blood mononuclear cells (PBMC) were obtained from the Shanghai Blood Center under a research protocol approved by the Department of Shanghai Blood Administration. PBMC were used either fresh or frozen in 10% dimethylsulphoxide (DMSO) containing fetal bovine serum (FBS). Frozen PBMC were thawed 1 day prior to the cultivation in RPMI-1640 medium supplemented with 10% fetal calf serum (FCS), 1% penicillin–streptomycin, 2 mM L-glutamine and 200 U/ml IL-2 in 5% CO2 at 37°C. MbIL-21-CD137L-K562 cells were pretreated with 15 μg/ml of mitomycin for 4 h and then washed twice with phosphate-buffered saline (PBS), mixed with PBMC at 2:1 and incubated in RPMI-1640 medium supplemented with 10% FCS, 1% penicillin–streptomycin, 2 mM L-glutamine and 100 U/ml IL-2 in 5% CO2 at 37°C. Repeated stimulation was performed weekly. For the STAT-3 inhibition experiment, JSI-124, a specific STAT-3 inhibitor, was added to a final concentration of 0·1 μM at the third stimulation, and DMSO was added as control. NK cell receptor expression, NK cell proliferation and cytotoxicity were analysed by flow cytometry, trypan blue staining and cytotoxicity assay at different time-points, respectively.
Flow cytometric analysis
Cells were exposed to appropriate antibodies for 30 min at 4°C, washed and resuspended in PBS containing 1% FBS. Data were acquired using a FACSCalibur cytometer (BD Biosciences) and analysed using FlowJo software (Ashland, OR, USA).
NK cell purification
Human peripheral blood mononuclear cells and red blood cells (RBC) were obtained from Shanghai blood centre under a research protocol approved by the Department of Shanghai Blood Administration. NK cells were purified using the RosetteSep Human NK Cell Enrichment Cocktail (StemCell Technologies, Vancouver, BC, Canada), as described previously . Briefly, 1 × 106 PBMC were mixed with 100 × 106 RBC before 1 μl RosetteSep reagent was added per 1 × 106 of PBMC. The cell mixture was incubated for 20 min at room temperature, and then diluted with an equal volume of RPMI-1640 followed by Ficoll-Paque isolation at 400 g for 20 min. Purified NK cells were used in subsequent experiments.
NK cell cytotoxicity was determined using the calcein release assay, a fluorometric assay comparable to the chromium release assay [8, 9]. Target K562 cells were labelled with 2 μg/ml calcein-AM for 1 h at 37°C with occasional shaking. Effector cells and target cells were co-cultured at the indicated effector-to-target (E : T) ratios and incubated at 37°C for 4 h. After incubation, 100 μl of the supernatant was transferred to a new plate. The fluorescence of the samples was measured with a Spectramax Gemini EM Fluorescence Microplate Reader (Molecular Devices, Sunnyvale, CA, USA) (excitation filter 485 nm, emission filter 538 nm). The percentage lysis was calculated according to the formula [(experimental release − spontaneous release)/(maximum release − spontaneous release)] × 100.
To investigate the effect of STAT-3 inhibitor JSI-124 on the viability of human NK cells, 1 × 106 primary purified or expanded NK cells were seeded per well in 24-well plates. JSI-124 was added at the indicated final concentrations (0, 0·05, 0·1, 0·2 and 0·5 μM). At the 24, 48 and 72 h time-points, cells were stained with 7-AAD, then analysed by flow cytometry.
Cell lysis and Western blot analysis
Primary NK cells were purified and incubated with 20 ng/ml of IL-21 with or without 0·1 μM of JSI-124 for 24 h, and were then lysed with 50 mM Tris-Cl (pH 6·8), 100 mM dithiothreitol, 2% sodium dodecyl sulphate (SDS) and 10% glycerol. Samples were analysed by SDS-polyacrylamide gel electrophoresis (PAGE), followed by immunoblotting using the Chemo Glow chemiluminescent substrate (Alpha Innotech, San Leandro, CA, USA) according to the manufacturer's instructions.
Results are expressed as the mean ± standard deviation. Statistical comparison was performed by Student's t-test. P-values of less than or equal to 0·05 were considered significant.
CD137L and mbIL-21 were expressed on the surface of K562 cells
We engineered K562 cells to express mbIL-21 and CD137L, and used these cells to expand NK cells efficiently from the peripheral blood mononuclear cells (Fig. 1). For cell engineering, CD137L and mbIL-21 sleeping beauty expression vectors were harvested as described in Materials and methods, and then transfected into K562 cells, together with the sleeping beauty transferase SB11. CD137L was first transfected, and CD137L-positive K562 cells (CD137L-K562) were sorted by the flow cytometer; mbIL-21 was transfected subsequently into CD137L-K562 cells, and mbIL-21-positive CD137L-K562 (mbIL-21-CD137L-K562) cells were sorted. Isolated cells were stained with CD137L and IL-21 flow cytometer antibodies. Results showed that both CD137L and IL-21 were expressed clearly on the surface of mbIL-21-CD137L-K562 cells (Supporting Fig. S1).
MbIL-21-CD137L-K562 cells elicited high-purity NK cells
After constructing the mbIL-21-CD137L-K562, NK cell expansion was performed as described in Materials and methods. To evaluate NK cell purity, expanded cells were stained with CD3, CD56 and CD16 antibodies. Figure 2 was a representative of six different expansions. As shown, CD3+ cells were approximately 60% and, initially, CD3−CD56+/CD16+ cells were less than 30% in PBMC; after a 1-week expansion, CD3+ cells were reduced to approximately 6%, and CD3−CD56+/CD16+ cells were increased to approximately 85%; after a 3-week expansion, CD3+ cells were decreased to less than 1%, and CD3−CD56+/CD16+ cells were increased to approximately 95%. These results showed that mbIL-21-CD137L-K562 cells induced the generation of high-purity human NK cells from peripheral blood mononuclear cells.
The expression of most NK cell-activating and inhibitory receptors was up-regulated
Besides CD56 and CD16, the NK cell surface has many other receptors, such as the activating receptors NKG2D, NKp30, NKp44, NKp46, NKp80, CD226 and 2B4, and the inhibitory receptors KIR2DL1, KIR2DL2 and KIR3DL1. The concerted action of these receptors determines NK cell lytic activity . Therefore, we analysed expression of the receptors on the expanded NK cell surface via flow cytometry. The results showed that other than the down-regulation of activating receptor NKp80, the expression of all other detected activating and inhibitory receptors were increased with the expansion (Fig. 3). In short, the data showed that expression of NK cell receptors were maintained, most of which were up-regulated during expansion.
MbIL-21-CD137L-K562 cells elicited high cytotoxicity and sustained proliferation of NK cells
Because balanced expression of NK cell receptors determines NK cell lytic activity, and both activating and inhibitory receptors (except for NKp80) were up-regulated in expanded NK cells, we evaluated the effectiveness of NK cell-mediated killing via cytotoxicity assay. The results showed that NK cell killing activity increased with expansion and reached the highest point at 3–5 weeks, then began to decrease after 6 weeks, although still significantly higher than unexpanded (resting) NK cells (Fig. 4a). These results showed that expanded NK cells were activated and functioned properly.
The goal of ex-vivo expansion was to produce large numbers of functional NK cells. As the expanded NK cells were functional, the next objective was to evaluate NK cell proliferation by counting the total cell numbers after trypan blue staining. The results showed that NK cells were increased significantly after expansion (Fig. 4b). Taken together, our results provide strong evidence showing that mbIL-21 could promote sustained NK cell proliferation and produce highly cytotoxic NK cells.
IL-21 activated but JSI-124 inactivated STAT-3
Because mbIL-21-CD137L-K562 induced large-scale and sustained proliferation of functional NK cells from peripheral blood mononuclear cells, we wanted to investigate the mechanisms involved. By screening the phosphorylation status of STAT-1–6 via Western blot, we found that only STAT-3 was phosphorylated continually in primary NK cells (unpublished data), which led us to hypothesize that STAT-3 activation is required for human NK cell maintenance and expansion. To test this hypothesis, we first examined the effect of IL-21 on STAT-3 phosphorylation in human NK cells. We discovered that IL-21 could increase STAT-3 phosphorylation, and the increase could be clearly reversed by STAT-3 inhibitor JSI-124 (Supporting Fig. S2A). These results support the hypothesis that IL-21 could activate STAT-3 in human NK cells, while JSI-124 could inhibit STAT-3 activation.
STAT-3 inhibition impaired NK cell proliferation and cytotoxicity
To study the effects of STAT-3 inhibition on NK cell proliferation and cytotoxicity, we first evaluated the toxicity of JSI-124 on primary and expanded NK cells and found that JSI-124 had no clear effect on NK cell viability in the concentrations tested (Supporting Fig. S2B). We then added a low dose of JSI-124 during NK cell expansion and discovered that JSI-124 could increase the population of CD3+ T cells and decrease the populations of CD16+, NKG2D+, NKp30+ and NKp44+ NK cells, while having no distinctive effect on other cell populations (Fig. 5). By comparing the mean expression levels of receptors induced by JSI-124 to those of the untreated control, we found that JSI-124 could decrease significantly the expression of most NK cell-activating and inhibitory receptors, except for NKp80 (Supporting Fig. S3). Moreover, we found that JSI-124 impaired normal NK cell morphology. Typically, NK cells were polymorphous after expansion; however, this morphology was lost with JSI-124 treatment (Fig. 6a). Further analysis showed that JSI-124 severely impaired NK cell proliferation (Fig. 6b) and cytotoxicity (Fig. 6c). Taken together, STAT-3 inhibition could impair NK cell morphology, receptor expression, cell proliferation and cytotoxicity. These results showed that STAT-3 activation is required for the mbIL-21-CD137L-K562-induced NK cell expansion ex vivo.
Adoptive NK cell transfer is a promising method to treat malignant tumours. However, this approach has been hampered by insufficient NK cells from donors. To overcome this limitation, novel methods to expand NK cells have been developed. In this study, we engineered a K562 cell line to directly express mbIL-21 and CD137L; with these cells, we generated large numbers of functional human NK cells from peripheral blood mononuclear cells, and discovered that NK cell expansion depends upon STAT-3 activation.
Functional NK cells could be expanded from purified NK cells [10, 11], umbilical cord blood cells [12, 13], haematopoietic stem cells  and PBMC [15, 16] by using cytokines, Epstein–Barr virus-transformed lymphoblastoid cells, heparin- and stromal cell-based cultures, and membrane-bound IL-15 and IL-21 artificial antigen present cells expressing CD64, CD86, CD19 and 4-1BBL  [18, 19]. All these methods provide an alternative approach for human NK cell ex-vivo expansion, but little was known about the NK cell expansion mechanism, which may benefit the design and development of human NK cell immunotherapy. In this study, by simply modifying the K562 cells to express mbIL-21 and CD137L, we developed an efficient method to expand functional human NK cells. More importantly, we discovered that STAT-3 activation is essential for NK expansion, suggesting that STAT-3 activation may benefit human NK cell proliferation and cytotoxicity. To confirm this speculation, we used a different cytokine of IL-10 to stimulate primary human NK cells, and found that IL-10 increased STAT-3 phosphorylation significantly and enhanced the expression of NK cell receptors and cytotoxicity; we also showed clear reverse effects with a STAT-3 inhibitor (unpublished data).
Contrary to an earlier report , we found in our study that STAT-3 phosphorylation could increase NK cell cytotoxicity. This inconsistency may come from species variation: we used human NK cells and the earlier study used murine NK cells and/or different cell applications: we used the expanded NK cells in vitro, while the earlier study used them to infiltrate tumour cells. Of course, additional experiments are necessary to test these hypotheses.
In conclusion, we developed a simple and efficient method to produce functional human NK cells from PBMCs, and discovered that STAT-3 phosphorylation is required for human NK cell proliferation and cytotoxicity. This may benefit the development of adoptive NK cell immunotherapy to treat viral diseases and cancers.
This work was supported by grants from National Natural Science Foundation (81071858; 81273216), Innovative Scientific Research Key Project of Shanghai Municipal Education Commission (11ZZ105), Leading Academic Discipline Project of Shanghai Municipal Education Commission (J50201) and Shanghai Key Laboratory of Tumor Microenvironment and Inflammation (11DZ2260200).