Abstract
Background/Aim: To obtain sufficient numbers of high-quality natural killer (NK) cells, we developed feeder cells using synthetic biology techniques. Materials and Methods: K562 cells were engineered to express membrane bound interleukin-2 (mbIL2) or interleukin-13 (mbIL13). Results: The incubation of human primary NK cells isolated from peripheral blood mononuclear cells (PBMCs) with these feeder cells significantly increased the number of activated NK cells compared to K562 parental cells. Fluorescence-activated cell sorting (FACS) analysis demonstrated that NKG2D activating receptors were abundant on the surface of NK cells expanded by K562-mbIL2 or mbIL13 cells. NK cells expanded on K562-mbIL2 or mbIL13 lysed cancer cells more effectively than those cultured with normal K562 cells. Using NK cells incubated with our feeder cells, we developed anti-CD19 chimeric antigen receptor (CAR)-NK cells. They showed robust cytotoxic effect against CD19 positive cancer cell line. Conclusion: Our newly developed feeder cells could provide useful tools for NK cell therapy.
NK cell-based immunotherapies are emerging as a promising approach for the treatment of tumors (1-3). NK cells constitute 5-15% of circulating lymphocytes in humans. They remove transformed or virus-infected cells through the production of granzymes and perforin without sensitization. NK cells recognize tumor cells by sensing a lack of MHC class I molecules or ligands that are mainly expressed on transformed cells. For this reason, many clinical trials using NK cells are on-going against blood and solid tumors (4-12). One of the biggest challenges in NK cell therapies was to obtain sufficient numbers of NK cells for clinical trials. Various methods have been developed to obtain activated NK cells. The use of feeder cells has been proven to greatly enhance the expression of activating receptors on NK cells, as well as the expansion of NK cells from PBMCs (13-17). In many clinical trials using NK cells, these feeder cells, including autologous PBMCs, tumor cell lines, and genetically engineered cells, are used (4, 6, 10, 12, 18-21). Expanded NK cells can be transduced with chimeric antigen receptors for the specific lysis of tumor cells (22).
Here, we report the invention of new feeder cells, which are K562 cells expressing membrane bound IL2 (K562-mbIL2) or IL13 (K562-mbIL13). IL13 is known to be closely related to IL4, which significantly up-regulates NK cell activity (23, 24). IL-2 has been known to stimulate and activate NK cell activity (25). Herein, NK cells co-cultured with these feeder cells showed robust expansion and exhibited potent lytic activity against cancer cells compared to parental K562 cells.
Materials and Methods
Production of genetically engineered K562 cells expressing membrane-bound interleukins. The interleukin sequences were identified at the NCBI nucleotide sites. PCR was performed in the following steps: 95°C denaturation (30 s), 62°C annealing temperature (30 s), and 72°C DNA extension (60 s). The PCR amplification was repeated for 30 cycles. We used Macrogen (Daejeon, Republic of Korea) CES sequencing service to confirm the inserted DNA sequences.
The K562 cell line was purchased from the American Type Cell Culture (ATCC, Manassas, VA, USA; CCL-243). The K562 cells were transduced with a lentiviral solution containing polybrene (hexadimethrine bromide) (Sigma-Aldrich, St. Louis, MO, US; H9268) by centrifugation at 1,500 × g for 90 min. Gene transduced K562 cells were selected with 4 μg/ml of puromycin for about a week.
Lentivirus production. On the first day, 7×106 293T cells were seeded in 10-mm tissue culture dishes. On day 2, transfer DNAs (12 μg) were co-transfected with lentiviral psPAX package DNA (5 μg) and pMD2.G-VSV-G envelop DNA (5 μg), using the calcium method. On the following day, the medium was replaced with DMEM/high glucose (Hyclone, Logan, UT, USA; SH30243.01) with 10% FBS. The lentivirus supernatants were harvested for 48 h and 72 h after transfection. Next, a 0.45-μm pore size polyethersulfone (PES) membrane syringe filter (Millipore, Burlington, MA, USA; SLHPR33RS) was used to filter the viral supernatant. The lentivirus supernatants were immediately frozen and stored at −70°C in a deep freezer until use.
Fluorescence-activated cell sorting assay. Live cells were centrifuged at 4,000 rpm for 5 min. The staining buffer was added to the cells and centrifuged at 4°C and 4,000 rpm for 5 min. The K562 cells were stained with a flag-tag antibody (Biolegend, San Diego, CA, USA; 637310) for 30 min to detect membrane-bound interleukins. To detect NK cell receptors, NK cells were stained with CD3-FITC (Biolegend; 300406), CD56-APC/Cyanine7 (Biolegend; 362512), and CD314-FITC (Biolegend; 320820). Next, 4% paraformaldehyde (Biosesang, Yongin-si, Republic of Korea; P2031) was used for cell fixation. Depending on the number of cells and pellet size, the suspension was prepared in 350-1,000 μl. FACS Canto instrument with BD FACSDiva Software (BD Biosciences, San Jose, CA, USA) was used for analysis.
NK cell isolation and cell culture. Human PBMCs were obtained from healthy donors who provided written informed consent according to protocols approved by Korea National Institute for Bioethics Policy (KoNIBP) Institutional Review Board (IRB; approval no. P01-201607-31-003). All experimental protocols using PBMCs were approved by IRB of KoNIBP. All methods using blood samples were performed in accordance with the institutional biosafety guidelines. PBMCs were separated by density-gradient centrifugation. K562 and K562-mbIL feeder cells were irradiated using a Biological X-ray Irradiator (150 Gy) (Precision, Madison, CT, USA; X-RAD 320) before use. In the case of using RPMI1640 medium (HyClone, Logan, UT, USA; SH30027.01) supplemented with IL-2 (PeproTech, Cranbury, NJ, USA; 200-02) with 10% FBS (Gibco, Carlsbad, CA, USA; 16000-044) for NK cell culture, PBMCs were co-cultured with feeder cells to expand the NK cells. In case of using NK Macs medium (Miltenyi Biotech, Bergisch Gladbach, Germany; 130-114-429) containing 5% human serum (Sigma-Aldrich; H3667), IL2 (PeproTech; 200-02), IL15 (PeproTech; 200-15) and IL21 (PeproTech; 200-21), NK cells were isolated from PBMCs using NK isolation kit (StemCell; Cambridge, MA, USA). Isolated NK cells were co-cultured with feeder cells. All cells were cultured at 37°C incubator in a 5% CO2 humidified atmosphere.
CAR NK cells were prepared by centrifuging NK cells with CAR retrovirus-containing polybrene at 1,800 × g for 90 min. After removing the supernatant, the CAR NK cells were cultured in NK Macs medium containing 5% human serum with IL2, IL15, and IL21 at 37°C and 5% CO2 in a humidified atmosphere.
Cell cytotoxicity assay. To measure cytotoxicity of NK cells, NK cells or CAR NK cells were co-cultured for 4 h or 24 h with NALM-6 cells in 96-well plates at various effector:tumor cell ratios. After co-culture, the same volume of Bright Glo™ (Promega, Madison, WI, USA; E2620) as the cell medium was added. The luminescence value was detected by EnSpire Alpha-Multimode Plate Reader machine (PerkinElmer, Waltham, MA, USA).
Statistical analysis. Statistical significance of differences was determined by t-test. Analyses were performed using Prism software (GraphPad Software). Statistical significance is indicated by *p<0.05; **p<0.01; ***p<0.001; ****p<0.0001.
Results
Generation of genetically engineered feeder cells. First, we developed constructs containing genes encoding membrane-bound IL-2 (mbIL2) and IL-13 (mbIL13) (Figure 1A and B). In these constructs, hinge and transmembrane domains were derived from CD8α. FLAG was incorporated between the interleukin and hinge domains for FACS analysis. These genes were cloned into lentiviral vectors to obtain viral particles for transducing mbIL2 or mbIL13 gene into K562 cells, followed by puromycin selection. FACS analysis demonstrated that the K562-mbIL2 and K562-mbIL13 cells successfully expressed mbIL2 and mbIL13 on their surface, respectively (Figure 1C).
Ex vivo expansion of NK cells by novel feeder cells. To evaluate the ability of these new feeder cells to induce proliferation of NK cells, isolated NK cells from PBMCs were co-cultured with feeder cells, including K562, K562-mbIL2, and K562-mbIL13. Five different PBMCs (PBMC1, PBMC2, PBMC3, PBMC4, and PBMC5) were used in this study. NK cells were isolated from each PBMC, then expanded on feeder cells. Proliferated NK cells were counted at indicated time points. These experiments were performed using different growth media, including RPMI1640 supplemented with 10% FBS and NK Macs medium supplemented with 5% human serum. NK cells from PBMCs 1 and 2 were cultured in the RPMI medium, and those from PBMCs 3, 4, and 5 were cultured in the NK Macs medium. Regardless of the medium or PBMCs, K562-mbIL2 and K562-mbIL13 cells were found to have a higher capacity to increase the number of NK cells than K562 parental cells (Figure 2).
Analysis of NKG2D expression in ex vivo expanded NK cells. To evaluate the induction of proliferation of NK cells by feeder cells, we performed FACS. NK cells from different PBMCs were proliferated in the presence of K562, K562-mbIL2, or K562-mbIL13 feeder cells. FACS data indicated that K562-mbIL2 and K562-mbIL13 cells increased CD3 negative/CD56 positive populations, which are regarded as natural killer cells, compared to K562 parental cells in 5 PBMCs out of 7 (Figure 3). NKG2D is one of the most important activating receptors of NK cells. Our FACS data indicated that NK cells fed with K562-mbIL2 and K562-mbIL13 expressed more NKG2D than NK cells fed with K562 parental cells in 5 PBMCs out of 7 (Figure 3). These results indicate that our feeder cells induced increased NKG2D expression in expanded NK cells.
Cytotoxicity of expanded NK cells against tumor cell line. We assessed the lytic activity of NK cells induced to proliferate by different kinds of feeder cells. We conducted this experiment with three different PBMCs (PBMC1, PBMC2, and PBMC3). NALM6 cells expressing luciferase were incubated with NK cells induced to proliferate in three different feeder cells for 4 h or 24 h. The data showed that NK cells induced to proliferate by K562-mbIL2 or K562-mbIL13 cells exhibited high lytic activity against NALM6 cells compared to K562 parental cells in all PBMCs (Figure 4). This indicates that our novel feeder cells produce more potent NK cells against tumor cells.
Development of anti-CD19 CAR-NK cells using new feeder cells. Using K562-mbIL2 cells, we developed CAR-NK cells targeting CD19. CAR constructs were designed as described in Figure 5A. Three different PBMCs were used to develop anti-CD19 CAR-NK cells. NK cells induced to proliferate by mbIL2 K562 cells were transduced with retroviral particles containing the CAR vector. We examined the lytic activity of anti-CD19 CAR-NK cells by incubating CD19 positive cancer cells, NALM6, with CAR NK cells. The lysis data indicated that anti-CD19 CAR-NK cells were specifically effective against CD19 positive cancer cells in all PBMCs (Figure 5B).
Discussion
NK cells have emerged as powerful anti-cancer agents, showing promising clinical results for hematological tumors (26-28). In preclinical and clinical models, the cytotoxic activity of NK cells in the body has been found to be associated with the occurrence of tumors (29, 30). Receptors on NK cells can recognize stressed, virus-infected, or transformed cells, such as tumor cells. At present, over 100 clinical trials using autologous or allogeneic NK cells are on-going (31). For the success of clinical trials using NK cells, it is critical to obtain a sufficient number of effective NK cells. Until now, there have been four sources of NK cells, namely peripheral blood, umbilical cord blood, induced pluripotent stem cells, and NK cell lines. Each source of NK cells has its own advantages and disadvantages. The NK-92 cell line simplifies the manufacturing process but needs to be irradiated before infusion into patients, which leads to poor in vivo persistence. Peripheral blood is a major source of NK cells used in clinical trials. Several strategies have been developed to proliferate NK cells from blood, including cytokine cocktails, feeder cells, membrane particles, and cytokine-antibody fusions (31, 32). It is unknown which interleukin is best suited for culturing NK cells (33, 34). The use of feeder cells is one of the most powerful methods for NK cell proliferation.
In the present study, we developed new feeder cell lines. In previous reports, membrane bound interleukins have shown a significant advantage in NK cell proliferation (35-38). NK cells expanded by genetically engineered feeder cells have shown promising results in clinical trials (7, 18, 19). In this study, we used IL-2 and IL-13 for membrane bound interleukin. IL-2 is a well-known interleukin critical for NK cell proliferation (39). IL-13 has many similarities to IL-4, which is critical for NK cell function (40). In NK cell proliferation test, K562-mbIL2 and K562-mbIL13 showed superior capacity compared to K562 parental cell. In this study, we cultured NK cells with two different media. One is RPMI1640 medium supplemented with FBS and the other is NK macs medium supplemented with human serum. Each medium has its own advantage for NK cell over others (41). In this study, NK macs medium supplemented with human serum was more favorable for NK cell proliferation than RPMI1640 medium with FBS. Our new feeder cells induced the marked expansion of NK cells in both media. For example, in AML, there may be more than 1012 leukemic cells at the time of diagnosis. To achieve a target-to-effector ratio of at least 1:100, 1010 NK cells are required. Therefore, it is important to develop a protocol for efficiently expanding the number of effective NK cells. To confirm our findings, we performed this test using five different PBMCs. The results were the same for all the PBMCs. In addition, NK cells fed with K562-mbIL2 or K562-mbIL13 exhibited higher NKG2D expression than the K562 parental cells. NKG2D is a receptor for MHC class I chain-related protein A (MICA), which is over-expressed in transformed cells. K562-mbIL2 and K562-mbIL13 showed more powerful lytic activity against NK cells. We postulated that the enhanced lytic activity of NK cells was due to the enhanced expression of NKG2D.
Using these new feeder cells, we developed anti-CD19 CAR NK cells. In a clinical study, NK cells engineered to express anti-CD19 CAR showed promising results (22). As shown in Figure 5, our anti-CD19 CAR NK cells showed excellent activity against CD19 positive cells. Therefore, we propose our NK cell culture system for the development of CAR NK therapy.
In this study, we developed new potent feeder cells for NK cells. Our data demonstrated that these new feeder cells efficiently expanded NK cells from peripheral blood. These new feeder cells could provide successful NK cell therapy for patients.
Acknowledgements
This work was supported by a grant from Korea Drug Development Fund (HN21C0226000021) and Korea Research Institute of Chemical Technology (No. KK2331-50).
Footnotes
Authors’ Contributions
C.H.P. conceived the experiments, analyzed the data, and wrote the manuscript. D.Y.L. designed the experiments. D.Y.L., Y.K., J.U.C., and J.S.P. performed the experiments. H.G.J. reviewed the manuscript and provided feedback.
Conflicts of Interest
The Authors have no conflicts of interest to declare in relation to this study.
- Received June 23, 2023.
- Revision received July 19, 2023.
- Accepted July 20, 2023.
- Copyright © 2023 International Institute of Anticancer Research (Dr. George J. Delinasios), All rights reserved.
This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY-NC-ND) 4.0 international license (https://creativecommons.org/licenses/by-nc-nd/4.0).