Cancer immunotherapy has received considerable attention in the last several decades. The most prominent advantage of immunotherapy over radiation and chemotherapy is its specific cytotoxicity against tumors without causing normal tissue damage. Adoptive immunotherapy aims to eliminate cancer cells through the transfer of ex vivo expanded and activated immune cells (Restifo et al. 2012). Immunologic effector cells, such as lymphokine-activated killer (LAK) cells, tumor-infiltrating lymphocytes (TIL), and cytokine-induced killer (CIK) cells, have been used in active immunotherapy of cancer (Guo and Han 2015).

Lymphokine-activated killer (LAK) cells are derived from human peripheral blood mononuclear cells and mouse splenocytes with high doses of IL-2 (Takei 2011; Restifo et al. 2012). LAK cells consist of a heterogeneous population of effector cells including CD3⁺CD56⁻ which can recognize tumor cells without major histocompatibility complex (MHC)-restriction (Dillman et al. 2009). However, LAK cell therapy may be interrupted by inherently low cytotoxic activity against tumor cells and the difficulty of generating a large number of cells (Schmidt-Wolf et al. 1991). Tumor-infiltrating lymphocytes (TIL) have stronger antitumor activity than LAK cells (Rosenberg et al. 1986). The major effectors of TIL cells are CD3⁺CD56⁻CD8⁺ and also exhibit MHC-restricted lysis of target tumor cells. However, TIL cell therapy is hindered by the difficulty of recovering the appropriate number of cells and the possibility of functional alteration during extraction from human tissues (Schmidt-Wolf et al. 1991; Lu and Negrin 1994; Schmidt-Wolf et al. 1997; Kim et al. 2015; Zhang et al. 2023).

In contrast, CIK cells are known to have several advantages. First, it is very easy to generate a larger number of CIK cells than LAK cells ex vivo (Kim et al. 2007a). Second, CIK cells have potent antitumor activity and exhibit almost no cytotoxicity toward normal hematopoiesis progenitor cells (Schmidt-Wolf et al. 1991; Scheffold et al. 1995; Li et al. 2010). Third, CIK cell-mediated cytotoxicity against target cells is MHC-unrestricted and T cell receptor-independent, with target killing occurring through NKG2D-mediated recognition (Laport et al. 2011). Finally, toxicity toward normal tissues is minimal and no graft-versus-host reaction is observed (Baker et al. 2001). CIK cells have shown promising antitumor effects against various cancers including hepatoma, leukemia, lung, ovarian, renal, and gastric cancers, in preclinical and clinical studies (Takayama et al. 2000; Wang et al. 2006; Kim et al. 2007b; Kim et al. 2010; Wongkajornsilp et al. 2013; Zhou et al. 2018).

Generation protocols of human CIK cells from human peripheral blood mononuclear cells (PBMC) are well established, but only limited information is available on mouse CIK cell generation. In the present study, we optimized the generation methods of large numbers of mouse CIK cells. We generated CIK cells from mouse spleen cells, characterized their phenotypes and cytokine gene expression profiles, and evaluated the antitumor activity of CIK cells in vitro.

Materials

Female C57BL/6 (H-2^b) mice (6-8 weeks old) were obtained from the Korea Research Institute of Bioscience and Biotechnology (Chungbuk, Korea). Mice were housed in specific pathogen-free conditions at 21-24℃ and 40-60% relative humidity under a 12-h light/dark cycle. All animals were acclimatized for at least 1 week before the experiments. The experimental procedures used in this study were approved by the Chungbuk National University Animal Experimentation Ethics Committee (Kim et al. 2010). Anti-mouse antibodies against CD3, CD4, CD8, NK1.1, and NKG2D were purchased from BD Biosciences (San Jose, CA, USA).

CIK cell generation

Single-cell suspensions were prepared from the spleens. One million cells were incubated in RPMI1640 medium containing 10% fetal bovine serum, 2 mM L-glutamine, 100 U/mL penicillin, 100 µg/mL streptomycin, and 50 µM 2-mercaptoethanol. The cells were incubated in a medium containing 1,000 U/mL of rmIFN-γ (R&D Systems, Minneapolis, MN, USA) for 24 h. Following this, the cells were transferred to dishes coated with 50 ng/mL of anti-CD3 antibody (17A2; R&D Systems, Minneapolis, MN, USA) and were incubated for 4 more days in a medium containing 250-750 U/mL of rmIL-2 (R&D Systems, Minneapolis, MN, USA). From day 5, the cells were sub-cultured every two days to maintain the cell density at approximately one-three million cells per milliliter. The viability of CIK cells was usually 85-95% (Lee et al. 2016).

Phenotype analysis

The phenotype of CIK cells was analyzed by flow cytometry. Cell staining was performed using CD3-FITC, CD4-FITC, CD8-PE, and NK1.1-APC. One million CIK cells were washed once with PBS containing 0.5% bovine serum albumin (BSA) and resuspended in 50 µL of PBS/BSA buffer. The cells were incubated with various conjugated monoclonal antibodies for 15 min at 4℃, washed twice with PBS, and resuspended in 500 µL of PBS (Kim et al. 2010). Flow cytometric analysis was performed on a FACSCalibur flow cytometer (BD Biosciences, San Jose, CA, USA), and the dates were analyzed using the WinMDI statistical software (Scripps, La Jolla, CA, USA). Forward and side scatter parameters were used to gate the live cells. Cell viability was examined by the propidium iodide (PI) nuclear staining method (Kim et al. 2009). The cells were stained with 1 µg/mL of PI and were analyzed with a FACSCalibur flow cytometer. The cells stained with PI were considered dead.

Reverse transcription-polymerase chain reaction (RT-PCR)

CIK cells were stimulated with 0.3-5 µg/mL of Concanavalin A (Con A) for 4 h. Total RNA was isolated using TRIZOL™ Reagent (Molecular Research Center, Cincinnati, OH, USA) (Kim et al. 2019). RT-PCR was performed to examine the changes in cytokine gene expression. For RT-PCR, single-stranded cDNA was synthesized from 3 µg total RNA. The primer sequences used were as follows: IL-2, sense, 5’-CTT GCC CAA GCA GGC CAC AG-3’, antisense, 5’-GAG CCT TAT GTG TTG TAA GC-3’; TNF-α, sense, 5’-AGG TTC TGT CCC TTT CAC TCA CTG-3’, antisense, 5’-AGA GAA CCT GGG AGT AGA CAA GGT A-3’; IL-4, sense, 5’-GAA TGT ACC AGG AGC CAT ATC-3’, antisense, 5’-CTC AGT ACT ACG AGT AAT CCA-3’; IFN-γ, sense, 5’-AGC GGC TGA CTG AAC TCA GAT TGT AG-3’, antisense, 5’-GTC ACA GTT TTC AGC TGT ATA GGG -3’; Perforin, sense, 5’-AGC CCC TGC ACA CAT TAC TG -3’, antisense, 5’-CCG GGG ATT GTT ATT GTT CC -3’; granzyme A, sense, 5’-ATT CCT GAA GGA GGC TGT GAA -3’, antisense, 5’-GCA GGA GTC CTT TCC ACC AC -3’; granzyme B, sense, 5’-GCC CAC AAC ATC AAA GAA CAG -3’, antisense, 5’-AAC CAG CCA CAT AGC ACA CAT -3’; Fas Ligand, sense, 5’-TAC CAC CGC CAT CAC AAC -3’, antisense, 5’-GAG ATC AGA GCG GTT CCA -3’; β-actin, sense, 5’- TGG AAT CCT GTG GCA TCC ATG AAA C-3’, and antisense 5’-TAA AAC GCA GCT CAG TAACAG TCC G-3’. PCR products were fractionated on 1% agarose gels and stained with 5 µg/mL ethidium bromide. After analyzing band areas using an image analysis system (Multi-Analyst, Bio-Rad, CA, USA), target mRNA expression levels were calculated as relative rations versus β-actin (Kim et al. 2008).

In vitro cytotoxicity assay

CIK cells were incubated with B16F10 cells, P815 cells, and Yac-1 cells in U-bottomed, 96-well plates at various effector-target ratios for 6 h at 37℃. Cytotoxicity was determined using a nonradioactive lactate dehydrogenase (LDH) release assay according to the manufacturer’s instructions. After 6 h of incubation, the plates were centrifuged and 50 µL of the supernatant was transferred to new 96-well flat-bottomed plates. The release of LDH into the supernatant was quantified by recording the absorbance at 490 nm. The percentage of specific lysis was calculated from LDH as follows: (experimental release − target spontaneous release − effector spontaneous release) / (target maximum release − target spontaneous release) × 100% (Lee et al. 2016).

In vivo anti-tumor activities of CIK cells

On day 0, B16F10 cells at 2 × 10⁵ cells/mouse were injected subcutaneously into C57BL/6 mice. CIK cells were injected intravenously once a week at doses of 1, 3, and 10 × 10⁶ cells per mouse. Adriamycin (ADR) was injected intravenously once a week at 2 mg/kg. The tumor volumes were estimated by the formula length (mm) × width (mm) × height (mm). On day 14, the mice were sacrificed and the tumor weights were measured. The body weights of the animals were also measured (Kim et al. 2010).

Statistics

Data represent mean ± SEM of three independent experiments and p-values were calculated using ANOVA software (GraphPad Software, San Diego, CA, USA).

Effect of IFN-γ, anti-CD3 Abs, and IL-2 on murine CIK cell generation

CIK cells were generated from mouse spleen cells by sequentially incubating them in a medium containing IFN-γ for 24 h, in anti-CD3 Ab-coated plates having IL-2-containing medium for 4 days, and by sub-culturing them every two days with IL-2-containing medium for 9 days (Fig. 1A). Under these culture conditions, the cells expanded rapidly to more than 200-fold by day 14 and the viability was approximately 80%. In the absence of initial IFN-γ stimulation, the number of cells expanded by approximately 100-fold on day 14, and the viability was approximately 80% (Fig. 1A). In addition, spleen cells stimulated with IL-2 alone or anti-CD3 Ab alone did not expand and not viable (Fig. 1A). We also examined the effect of IL-2 dosages on CIK cell generation. Upon addition of IL-2 at 250 U/mL, 500 U/mL, and 750 U/mL to the culture medium, CIK cells expanded 7-fold, 48-fold, and 164-fold, and the viabilities were 39%, 62%, and 75%, respectively (Fig. 1B). FACS analysis was performed to characterize the subsets of CIK cell population. The phenotypes of fresh spleen cells were 30% CD3⁺NK1.1⁻, 5% CD3⁻NK1.1⁺, and 2% CD3⁺NK1.1⁺ (Fig. 2A). After being cultured for 14 days, the population was changed to 52% CD3⁺NK1.1⁻, < 1% CD3⁻NK1.1⁺, and 45% CD3⁺NK1.1⁺ (Fig. 2B). In addition, most CD3⁺NK1.1⁺ cells were CD8⁺, but not CD4⁺. These results suggest that three activators, namely IFN-γ, anti-CD3 Abs, and IL-2, are essentially required to generate a large enough number of murine CIK cells including CD3⁺NK1.1⁺ cells.

Figure 1. Effect of cytokines and anti-CD3 antibodies on CIK cell generation. (A) Mouse spleen cells were cultured with anti-CD3 antibodies alone, IL-2 alone, anti-CD3 antibodies plus IL-2, or anti-CD3 Abs plus a cytokine cocktail (IFN-γ and IL-2). (B) IL-2 was used at concentrations of 250, 500, and 750 U/mL. The cell number was counted with a hemacytometer, and viability was determined by trypan blue dye exclusion assay.

Figure 2. The subsets of CIK cell population. Mouse spleen cells were cultured with IFN-γ, immobilized anti-CD3 antibodies, and 750 U/mL IL-2 for 14 days. Fresh spleen cells (A) and CIK cells (B) were stained with antibodies, such as anti-CD3-FITC plus anti-NK1.1-PE, anti-CD4-FITC plus anti-CD8-PE, anti-CD4-FITC plus anti-NK1.1-PE, and anti-CD8-FITC plus anti-NK1.1-PE. The cells were stained with PI to examine the viability.

Effect of cell density on CIK cell expansion

During the generation periods of CIK cells, proliferation, and viability were assessed every day. During 14 days, the cells showed biphasic expansion; they slowly expanded 2-3-fold until day 4, but rapidly from day 5 to day 14. We found that frequent subculture from day 5 was crucial for cell expansion. When the cell number was adjusted to 1-3 × 10⁶ cells/mL, the total cell number increased up to 200-fold on day 14 (Fig. 3A) and the viability was maintained at approximately 80% (Fig. 3B). However, when we missed adjusting the cell density to 1-3 × 10⁶ cells/mL from day 4, most cells died within 24 h, which might be due to contact inhibition (Fig. 3B).

Figure 3. Effect of cell density on CIK cell generation. Mouse spleen cells were sequentially cultured with IFN-γ, immobilized anti-CD3 antibodies, and IL-2 for 14 days. Subculture was performed every two days from day 4 in order to adjust the cell density to 1-3 × 10⁶ cells/mL. The cell number was enumerated with a hemacytometer (A), and the cell viability was determined by trypan blue exclusion assay (B).

Cytokine and cytolytic mediator expression of CIK cells

Next, we characterized the functional properties of CIK cells generated from spleen cells. We examined the cytokine expression profiles of CIK cells. CIK cells, harvested on day 14, were re-activated in vitro with 0.3 to 3 µg/mL of Concanavalin A (Con A) for 4 h. RT-PCR analysis was performed using RNA extracted from CIK cells. Although TNF-α and IL-4 were weakly expressed, CIK cells strongly expressed mRNA of IFN-γ and IL-2 (Fig. 4A). We also investigated the gene expression levels of cytolytic mediators. CIK cells highly expressed Fas ligand (FasL) and granzyme A/B, but weakly perforin (Fig. 4B).

Figure 4. Expression profiles of cytokines and cytolytic mediators by CIK cells. The CIK cells and purified T cells were stimulated with Concanavalin A (Con A) for 4 h. The gene expression levels of cytokines, such as IFN-γ, TNF-α, IL-2, and IL-4 were analyzed by RT-PCR (A). The gene expression levels of cytolytic mediators, such as granzyme A/B, perforin, and FasL, were analyzed by RT-PCR (B).

Anti-tumor activity of CIK cells

We evaluated the cytotoxic lysis of tumor cells by CIK cells. Cytotoxicity of CIK cells against cancer cell lines, such as B16F10, P815, and Yac-1, was assessed by the LDH assay. When target cells were co-incubated with CIK cells for 6 h, CIK cells appeared to kill the target cells (Fig. 5A). However, fresh T cells did not destroy the target cells (Fig. 5B). The in vivo anti-tumor effects of CIK cells were also studied using a syngeneic tumor model with B16F10 cells. B16F10 cells at 2 × 10⁵ cells/mouse were injected s.c. and they grew up to 422 ± 100 mm³ of the tumor volume in 14 days. CIK cells, which were injected i.v. once a week at doses of 1, 3, or 10 × 10⁶ cells/mouse, inhibited the in vivo tumor growth by 23%, 50%, and 68%, respectively (Fig. 6A). On day 14, the tumors were isolated and weighed, and they demonstrated a strong anti-tumor effect of CIK cells (Fig. 6B and 6C). The body weights of mice were examined to assess the toxicity of CIK cells, which showed that CIK cells did not induce weight loss (Fig. 6D).

Figure 5. In vitro cytotoxic activity of CIK cells. In vitro cytotoxicity of CIK cells was assessed against B16F10, P815, and Yac-1 target cells. Target cells were incubated with either CIK cells (A) or purified T cells (B) for 6 h at effector-to-target (E:T) ratios of 3-30:1. Cytotoxic activity was measured by LDH cytotoxicity assay.

Figure 6. In vivo antitumor activity of CIK cells. B16F10 cells were injected s.c. to mice and CIK cells were injected i.v. every week. ADR was injected intravenously every week at 2 mg/kg. The tumor volume was calculated using the formula: length (mm) × width (mm) × height (mm) (A). On day 14, the tumors were isolated and weighed (B). Photographs of isolated tumors (C). The body weights were measured (D). *p < 0.01. **p < 0.001.

Cancer immunotherapy aims to eliminate cancer cells and to improve the quality of life of patients with tumors. CIK cell-based immunotherapy has recently received much attention (Feng et al. 2022). CIK cells show anti-tumor activity against various cancers in vitro and in vivo (Wang et al. 2002; Zhao et al. 2010; Kim et al. 2012). In the present study, we provided the optimal information to generate CIK cells from mouse spleen cells with three major activators, IFN-γ, IL-2, and anti-CD3 antibody.

Two critical factors were considered to successfully generate CIK cells. First, CIK cells could be generated by stimulation with both anti-CD3 antibody and IL-2, but not with each alone. Anti-CD3 antibody as a T cell activator is known to trigger the proliferation of T cells (Schmidt-Wolf et al. 1991). IL-2 is necessary for the proliferation, survival, and functioning of T cells. The interaction of IL-2 and IL-2 receptors stimulates the growth, differentiation, and survival of cytotoxic T cells (Beadling and Smith 2002). IL-2 is a growth factor for cytotoxic T cells to enter the S phase of the cell cycle and induced STAT5 signaling increases the expression of the antiapoptotic gene bcl-2 and cyclins, which are necessary for cell cycle progression (Janas et al. 2005). Also, IL-2 can initiate the expression of IFN-γ by T and NK cells (Farrar et al. 1981). We showed that a high enough concentration of IL-2 was required for the successful proliferation and survival of CIK cells. Of interest was the usage of IFN-γ for CIK cell generation. IFN-γ has been known to increase the cytotoxicity of CIK cells when it is added on day 0 before anti-CD3 antibody activation, whereas the reverse order leads to a decrease in cytotoxicity (Schmidt-Wolf et al. 1991). For these reasons, we treated spleen cells with IFN-γ on day 0. Interestingly, we found that IFN-γ was required to increase CIK cell proliferation, but had no effect on cell viability or cell phenotypes, suggesting that IFN-γ may play a key role in the induction of IL-2 receptor expression on CIK precursor cells (Itoh et al. 1985). These data suggest that successful generation of CIK cells from mouse spleen cells depends on the sequential activation of spleen cells with IFN-γ alone for one day, with anti-CD3 Abs and IL-2 for 4 days and with IL-2 alone for 9 more days.

The second critical factor for CIK cell generation was the adjustment of cell density during the culture periods. We observed that CIK cells underwent death during culture when the cell density rose above 3 × 10⁶ cells/mL. Cell density has been reported to be an important factor in maintaining certain T cell viability in vitro (Li et al. 2010). Cell growth in an in vitro culture system is arrested at high cell density termed contact inhibition (Seluanov et al. 2009). The high cell density of CIK cells during the generation period puts stress on cells. This stress is referred to as density-dependent inhibition of growth. Therefore, in the present study, we can presume that the high cell density of CIK cells induced cell death via density-dependent inhibition of growth. In addition, we showed that cell numbers should be maintained in the range from one to three million cells per milliliter. Otherwise, most cells died during cultivation. When we first cultured spleen cells at a concentration of 1 × 10⁶ cells/mL with our protocol, the cell number increased 3-4 fold up to day 4, and then, the cell number increased dramatically with an average 2-fold per 1 or 2 days. This suggested that subculture should be done frequently from day 5 to adjust the cell density to 1-3 × 10⁶ cells/mL.

Finally, we verified that CIK cells generated from spleen cells had typical characteristics of CIK cells generated from human PBMC. On day 14, the absolute cell number increased by approximately 200-fold and the phenotypes of the cell population were 97% CD3⁺, < 1% CD3⁻NK1.1⁺, 45% CD3⁺NK1.1⁺, 5% CD4⁺, 1% CD4⁺NK1.1⁺, 88% CD8⁺, and 32% CD8⁺NK1.1⁺, which were typical characteristics of human CIK cells. CIK cells showed typical cytokine expression profiles, in that they expressed IFN-γ, TNF-α, and IL-2, but not IL-4. It was interesting to note that CIK cells produced significant amounts of IFN-γ, which could signify the relationship with attenuated graft-versus-host-disease (GVHD). Allogenic spleen cells with bone marrow cells could induce GVHD in recipient mice by demonstrating body weight loss and death within 9 days after transplantation (Janas et al. 2005). However, CIK cells did not induce GVHD, even when mice were injected with 10 times the amount of CIK cells compared to that of spleen cells (Baker et al. 2001). This reduction in GVHD was known to be related to IFN-γ production of CIK cells. IFN-γ^+/+ CIK cells produced no GVHD, but IFN-γ^−/− CIK cells caused acute lethal GVHD (Baker et al. 2001; Leemhuis et al. 2005; Kim et al. 2007a). Another important role of IFN-γ released from CIK cells is modulation of the expression of adhesion molecules on target cells. IFN-γ could play an important role as an inducer of CD54 expression in primary chronic lymphocytic leukemia. This corresponded to an increase in contact between CIK and chronic lymphocytic leukemia, which was followed by the enhancement of CIK cell-mediated apoptosis (Kornacker et al. 2006).

In summary, we demonstrated here that CIK cells could be generated from mouse spleen cells, which may depend on two critical factors; sequential activation with three activators, namely IFN-γ, anti-CD3 antibodies, and IL-2, and adjustment of cell density to 1-3 × 10⁶ cells/mL during incubation. Mouse CIK cells have similar characteristics to human CIK cells in that they are highly proliferative by more than 200-fold, contained > 30% CD3⁺NK1.1⁺, produced high levels of IFN-γ, and had strong antitumor activity in vitro and in vivo. These data might be helpful for researchers, who want to use mouse CIK cells in their preclinical studies. For example, a study of in vivo distribution of CIK cells requires i.v. injection of mouse CIK cells into syngeneic recipient mice to match the species variation of adhesion molecules and chemokine receptors expressed in CIK cells and their ligands expressed by endothelial cells.

The authors declare no potential conflicts of interest.

This study was supported by grants funded by the Korean Government (NRF-2017R1A5A2015541).