Mouse CD8+NKT-like cells exert dual cytotoxicity against mouse tumor cells and myeloid-derived suppressor cells
Our previous work has demonstrated the high efficiency of CD8+ natural killer T (NKT)-like cells in killing antigen-bearing dendritic cells. To evaluate their role in the tumor microenvironment, we performed in vitro and in vivo antitumor experiments to investigate whether CD8+NKT-like cells could kill Yac-1 and B16 cells like NK cells and kill EL4-OVA8 cells in an antigen-specific manner like cytotoxic T lymphocytes (CTLs). Unlike NK1.1−CTLs, CD8+NKT-like cells also exhibit the capability to kill myeloid-derived suppressor cells (MDSCs) in an antigen-specific manner, indicative of their potential role in clearing tumor antigen-bearing MDSCs to improve the antitumor microenvironment. In vitro blocking experiments showed that granzyme B inhibitor efficiently suppressed the cytotoxicity of CD8+NKT-like cells against tumor cells and MDSCs, while Fas ligand (FasL) or tumor necrosis factor-related apoptosis-inducing ligand (TRAIL) inhibition failed to produce similar effects. Transcriptomic and phenotypic analyses of CD8+NKT-like cells, NK cells, and NK1.1−CTLs indicated that CD8+NKT-like cells expressed both T-cell activation markers and NK cell markers, thus bearing features of both the activated T cells and NK cells. Taken together, CD8+NKT-like cells could exert NK- and CTL-like antitumor effects through the elimination of both tumor cells and MDSCs in a granzyme B-dependent manner.
KeywordsCD8+NKT-like cells Antitumor effects MDSC Cytotoxicity Granzyme B
American Type Culture Collection
GFP-expressing B16 cells
- CD8+NKT-like cells
CD8-expressing natural killer T-like cells
Cytotoxic T lymphocytes
EL4 cells transfected with OVA257–264 peptide
EL4 cells transfected with both OVA257–264 peptide and GFP
Granulocyte–macrophage colony-stimulating factor
- iNKT cells
Invariant natural killer T cells
Killer cell lectin-like receptor G1
Myeloid-derived suppressor cells
NK1.1 negative cytotoxic T lymphocytes
Natural killer group 2D
- NKT cells
Natural killer T cells
β2 microglobulin deficient
Natural killer T (NKT) cells, as a population of lymphocytes bearing both T and NK cell lineage markers , have been the focus of immunological studies for decades [2, 3]. Based on their CD1d dependency and α-GalCer reactivity, NKT cells can be divided into type I (invariant NKT [iNKT] cells), type II, and NKT-like cells [1, 4]. Development of Jα18−/− mice  and CD1d tetramer  has encouraged extensive investigation of the antitumor effects by iNKT cells in both basic research and clinical trials [2, 7]. Immunologists have demonstrated that the activation of iNKT cells efficiently inhibits melanoma, thymoma, and sarcoma both in vitro and in vivo [8, 9, 10] in an interferon (IFN)-γ-dependent manner [11, 12]. An iNKT cell agonist, α-GalCer, has been recently employed in clinical practices to enhance the antitumor effects in patients with cancer [13, 14, 15]. Unlike iNKT cells, type II NKT cells are CD1d restricted but express a relatively diverse T-cell receptor (TCR) repertoire .
In contrast, NKT-like cells are CD1d independent and express diverse TCR repertoire, indicative of their ability to recognize antigens in a manner similar to that of conventional T cells . NKT-like cells from β2m−/− mice exhibited high cytotoxic effects on tumor cells in vitro . NK1.1+CD8 T cells from OT-I mice performed rapid and vigorous killing of tumor cells, while the NK1.1− cytotoxic T lymphocyte (CTL) population failed to exhibit potent antitumor effects, indicative of the more efficient tumoricidal effects of CD8+NKT-like cells . Although the antitumor potential of these CD8+NKT-like cells has been proposed, a detailed understanding of the role that CD8+NKT-like cells play is insufficient.
Our previous study has suggested that CD8+NKT-like cells could efficiently kill antigen-bearing dendritic cells (DCs) . Therefore, we speculate that since CD8+NKT-like cells have a high cytotoxic capability, they may also kill tumor cells. Herein, we performed in vitro and in vivo assays to demonstrate that CD8+NKT-like cells exert cytotoxicity against tumor cells in both a NK-like and a CTL-like manner. As CD8+NKT-like cells could kill antigen-bearing DCs, we investigated whether these cells could eliminate myeloid-derived suppressor cells (MDSCs) in the tumor microenvironment. As expected, we found that CD8+NKT-like cells could also kill MDSCs in an antigen-specific manner, suggestive of an alternative antitumor mechanism of CD8+NKT-like cells. Further investigation of the mechanism underlying the effects of these cells on tumor cell and MDSC killing was carried out and their NK- and CTL-like cytotoxic capabilities were evaluated to indicate the distinct immunological features of CD8+NKT-like cells.
Materials and methods
Cell lines and reagents
Murine Yac-1, B16-F10, EL4 cells as well as ovalbumin (OVA)257–264 peptide-expressing EL4-OVA8 cells and GFP-expressing B16-GFP and EL4-OVA8-GFP cells were cultured in RPMI-1640 medium (Gibco, MA, USA) supplemented with 10% FCS (Gibco, MA USA). Recombinant mouse cytokines, including granulocyte–macrophage colony-stimulating factor (GM-CSF), interleukin (IL)-2, IL-4, IL-6, IL-7, and IL-15, were obtained from PeproTech (NJ, USA). The fluorescent dyes used for cell staining were 5-chloromethylfluorescein diacetate (CMFDA), Hoechst 33,342, and 7-aminoactinomycin D (7-AAD; Life Technologies, MA, USA).
Wild-type (C57BL/6 mice) and transgenic mouse strains, including mT/mG mice (B6.129(Cg)-Gt(ROSA)26Sortm4(ACTB-tdTomato,-EGFP)Luo/J mice), B6.GFP mice (C57BL/6-Tg(ACTB-EGFP)1Osb/J mice), OT-I mice (C57BL/6-Tg(TcraTcrb)1100Mjb/J mice) and Simian virus 40 (SV40)-specific TCR transgenic mice (B6.Cg-Tg(TcraY1,TcrbY1)416Tev/J mice), were used for experiments at 6–10 weeks of age. To generate tdTomato fluorescence transgenic OT-I mice, female mT/mG mice were mated with male OT-I mice and their specific offspring were identified.
Flow cytometric analysis
To analyze the lymphocyte subsets and detect their surface marker expression, the lymphocytes were stained with fluorescent antibodies against TCRβ, NK1.1, CD49b, CD8, CD25, CD27, CD44, CD62L, CD69, CD107a, CD122, CD132, CD335, killer cell lectin-like receptor G1 (KLRG1), natural killer group 2D (NKG2D), Ly49D, Ly49G2, Ly49H, NKG2A/C/E, and various types of TCRs (BD Pharmingen, CA, USA). Expression of granzyme B and perforin was detected following intracellular cytokine staining protocols. Flow cytometry was performed using a FACSAria II (Becton–Dickinson, CA, USA), and the data were analyzed using FlowJo software.
Isolation and activation of CD8+NKT-like cells, NK cells, and NK1.1−CTLs
Splenocytes from OT-I mice (or SV40-specific TCR transgenic mice) were cultured in RPMI-1640 medium supplemented with 10% FCS, 50 ng/mL recombinant mouse IL-2, 10 ng/mL recombinant mouse IL-7, 50 ng/mL recombinant mouse IL-15, and DC-loaded OT-1 peptides (at 1 × 104 per 106 splenocytes) for 7 days. These lymphocytes were collected and stained with fluorescent antibodies. The population of CD8+NK1.1+TCRβ+ (CD8+NKT-like cells), NK1.1+TCRβ− (NK cells) and CD8+NK1.1−TCRβ+ (NK1.1−CTLs) cells was sorted using FACSAria II flow cytometry.
CD8+NKT-like cells, NK cells, and NK1.1−CTLs were isolated as described above and high-throughput transcriptome sequencing was performed at Beijing Genomics Institute. Relative intensities of all genes among CD8+NKT-like cells, NK cells, and NK1.1−CTLs were plotted as heat maps to depict their relationship. Expression profiles of the markers of T-cell activation, adhesion, and cytotoxicity as well as NK cell receptors were also shown as heatmaps.
B16-F10 cells were stained with DDAO-SE (red) and co-cultured with CD8+NKT-like cells on slides. Fixation and permeabilization were carefully carried out using intracellular cytokine staining kit (BD Pharmingen, CA, USA) on slides at 8, 16, and 24 h. These cells were stained with Hoechst 33,342 (blue, Life Technologies, MA, USA) and phycoerythrin (PE)-conjugated antibody against granzyme B (red). Images demonstrating the interaction between B16-F10 and CD8+NKT-like cells were obtained by BD Pathway 855 High Content Imager.
Isolation of MDSCs
Splenocytes from mT/mG mice were isolated and stained with allophycocyanin-conjugated antibody to CD11b and PE-conjugated antibody to Gr-1. CD11b+Gr-1+ cells were sorted as MDSCs.
Live cell imaging
Effector cells (CD8+NKT-like cells, NK cells, or NK1.1−CTLs) from tdTomato fluorescence transgenic OT-I mice were co-cultured with B16-GFP, EL4-OVA8-GFP, or MDSCs loaded with 1 μg/mL OVA257–264 peptides at E:T ratios of 1:3, 1:5, or 1:10, respectively, on confocal dishes. A dynamic process showing the interaction between the effector cells (red) and target cells (green) was recorded using Andor spinning disk live cell confocal microscopy with a 40 × oil immersion lens.
In vitro cytotoxicity assay
Target tumor cells were stained at 4 °C with CMFDA (Molecular Probes, Invitrogen) at a concentration of 1 μmol/mL for 106 cells. After 10 min of incubation, cells were washed thrice with PBS containing 10% FCS. The effector cells were co-cultured in 96-well plates with 1 × 104 target cells in RPMI-1640 containing 10 % FCS and 50 ng/mL of recombinant IL-2 at indicated E/T ratios. Cells were harvested every 12 or 24 h and incubated with 7-AAD (Molecular Probes, Invitrogen) at room temperature for 10 min. The cells were washed once with PBS and analyzed on BD FACSAria II. The percentage of 7-AAD-positive cells indicated the killing rate.
In vivo adoptive transfer assay
A total of 5 × 104 B16 melanoma cells or 5 × 106 EL4-OVA8 thymoma cells were intravenously or subcutaneously inoculated into recipient C57BL/6 mice, respectively. Effector CD8+NKT-like cells, NK cells or NK1.1−CTLs were used for peritoneal adoptive transfer. The tumor growth and survival rates were followed and recorded at the indicated time points or at the end of experiments.
Examination of tumor antigen-loaded MDSCs
A total of 2 × 106 EL4 or EL4-OVA8 cells were subcutaneously injected. On day 10 when the tumor size reached about 1 cm3, mice were sacrificed and tumors were resected. The tumors were digested with 1 mg/mL of collagenase IV (Sigma) at 37 °C for 1 h. Dissociated cells were collected through a 70 μm filter and stained with allophycocyanin/Cy7-conjugated anti-CD45.2, peridinin chlorophyll protein complex (PerCP)-conjugated anti-CD11b, PE-conjugated anti-Gr-1, and allophycocyanin-conjugated anti-H2Kb bound to SIINFEKL (which could recognize the OVA257–264–H2Kb complex). In this assay, allophycocyanin-conjugated mouse IgG1, a κ-isotype control antibody (BioLegend), was used as an isotype control. Intratumoral MDSCs were identified as CD11b+Gr-1+ cells and their expression level of OVA257–264–H2Kb complex was evaluated.
A two-tailed Student’s t test was used to compare two groups of normally distributed data and a Mann–Whitney U test was used when data were non-normally distributed. Error bars show standard errors. Difference between groups was considered statistically significant at P < 0.05 or less. ***P < 0.001, **P < 0.01, *P < 0.05, and “ns” indicated not significant.
CD8+NKT-like cells kill tumor cell-like NK cells
CD8+NKT-like cells kill tumor cells in an antigen-specific manner
CD8+NKT-like cells kill tumor antigen-bearing MDSCs
CD8+NKT-like cells exert cytotoxicity via a granule exocytosis pathway
Immunological features of CD8+NKT-like cells
NK cells and CTLs are the two most widely studied lymphocyte subsets in cancer immunotherapy [19, 20, 21]. Tumor antigens could be cross-presented to CTLs in an MHC-I-dependent manner [22, 23, 24], resulting in the activation of antigen-specific CTLs and efficient killing of tumor cells. However, under some circumstances, tumor cells may escape from the CTL-mediated elimination due to abnormal expression of MHC-I, mutation in tumor antigen, or MDSCs [25, 26, 27]. Unlike CTLs, NK cells could distinguish tumor cells from normal cells by recognizing the abnormal expression of MHC-I molecules or other NK cell receptor ligands [20, 21], thereby complementing the function of CTLs. Since the 1980s, adoptive cell transfer using NK cells and CTLs has been exploited in clinical trials, although with limited success [28, 29, 30, 31]. Thus, the development of a novel technology or concept to treat cancer is needed.
The roles of CD8+NKT-like cells in tumor immunity are unclear because of their scarcity . In our previous work, we established an in vitro method to amplify the population of CD8+NKT-like cells to facilitate a more comprehensive investigation of the functions of these cells . In this context, we showed that CD8+NKT-like cells exerted potent cytotoxicity against tumor cells and MDSCs and consequently inhibited the growth of tumors. Yac-1 is an NK cell-sensitive cell line that was also found to be killed by CD8+NKT-like cells almost as efficiently as by NK cells; thus, CD8+NKT-like cells exert an antigen-independent NK-like cytotoxicity. The high expression of NKG2D on CD8+NKT-like cells may be responsible for this effect, as NKG2D was shown to be involved in the NK cell-mediated killing of Yac-1 cells [32, 33]. In addition, our data also show that these CD8+NKT-like cells kill tumor cells in an antigen-specific CTL-like manner. However, the CD8+NKT-like cell-mediated cytotoxicity against OVA-expressing tumor cells was higher than the NK1.1−CTL-mediated cytotoxicity. This phenomenon has been previously reported . Thus, our data showed that CD8+NKT-like cells could kill tumor cells in an antigen-independent NK-like and an antigen-specific CTL-like manner.
The number of CD8+NKT-like cells is much smaller than that of NK cells or NK1.1−CTLs in naïve mice, but we demonstrated that antigen activated CD8+NKT-like cell responses lagged behind the response of NK1.1−CTLs in our previous work . Therefore, although in the late contraction stage of an immune response, a considerable number of CD8+NKT-like cells was still maintained and started to take over the roles of CTLs in immune responses and homeostatic maintenance . Based on this observation, we suggest that distinct from NK cells and NK1.1−CTLs that act as the first line of defense against tumors, the small population of CD8+NKT-like cells with more powerful antitumor capabilities may function as the second line of defense. Thus, it is also reasonable to suggest that the adoptive transfer of these CD8+NKT-like cells may provide efficient protection against tumors, as described in this context.
Myeloid-derived suppressor cells are considered as immunosuppressive lymphocytes in many physical and pathogenic processes and promote the escape of cancer cells from immune attacks [34, 35]. Some groups have developed approaches to eliminate or modulate the functions of these cells in vivo ; however, they remain resistant to cell-mediated lysis. Herein, we found that CD8+NKT-like cells exerted cytotoxicity against MDSCs in an antigen-specific manner, which has the potential to restore antitumor immunity. This observation was initially confusing with regard to our previously published study showing that CD8+NKT-like cells can kill antigen-bearing DCs to reduce inflammation. However, given the chronic inflammatory state often associated with tumorigenesis [37, 38], our findings are consistent and show that CD8+NKT-like cell-mediated cytotoxicity can regulate inflammation and restore homeostasis.
We thank Prof. Ying Wan (Army Medical University, Chongqing, China) for his help with GFP and OVA transfected cells.
MZ conceived the concept. MZ and CW designed the experiments. ZL and YW performed the in vitro and in vivo experiments. MZ, CW and ZL analyzed the data. MZ and CW wrote the paper.
This work was supported by the National Natural Science Foundation of China (81272321, 81771687, 81571532 and 81601429). We are grateful for the funding as it provided the necessary reagents and materials for this work.
Compliance with ethical standards
Conflict of interest
The authors declare that they have no potential conflicts of interest.
Ethical approval and ethical standards
All mouse experiments were conducted in accordance with the “Guide for the Care and Use of Laboratory Animals” (US Department of Health and Human Services, National Institute of Health Publication no 93–23, revised 1985) and were approved by the Institutional Animal Care and Use Committee, Tsinghua University, Beijing, China. The animal research approval number is 15-ZMH1.
Both wild-type (C57BL/6 mice) and transgenic mouse strains including mT/mG mice (B6.129(Cg)-Gt(ROSA)26Sortm4(ACTB-tdTomato,-EGFP)Luo/J mice), B6.GFP mice (C57BL/6-Tg(ACTB-EGFP)1Osb/J mice), OT-I mice (C57BL/6-Tg(TcraTcrb)1100Mjb/J mice), and Simian virus 40 (SV40)-specific TCR transgenic mice (B6.Cg-Tg(TcraY1,TcrbY1)416Tev/J mice) were purchased from the Jackson Laboratory (Bar Harbor, ME, USA) and bred in specific pathogen-free conditions in the Laboratory Animal Research Center, Tsinghua University (Beijing, China).
Cell line authentication
Murine Yac-1, B16-F10, and EL4 cells were obtained from ATCC (Manassas, VA, USA) and the STR profiles (Supplementary Fig. 2) of these murine cell lines, including 18-3, 9-2, 6-7, 5-5, X-1, 15-3, 12-1, 6-4, CSF1PO, vWA, 4-2 and Jarid1 loci, were detected in Beijing Microread Genetics Co., Ltd (Beijing, China). Transfection of EL4 and B16-F10 cells, for OVA257–264 peptide-expressing (EL4-OVA8) cells and GFP-expressing cells (B16-GFP and EL4-OVA8-GFP), was performed in Prof. Ying Wan’s laboratory. Expressions of transfected GFP/OVA sequences were validated by flow cytometry (Supplementary Figure 3).
- 1.Godfrey DI, Hammond KJ, Poulton LD, Smyth MJ, Baxter AG (2000) NKT cells: facts, functions and fallacies. Immunol Today 21(11):573–583Google Scholar
- 2.Bendelac A, Savage PB, Teyton L (2007) The biology of NKT cells. Annu Rev Immunol 25:297–336. https://doi.org/10.1146/annurev.immunol.25.022106.141711 Google Scholar
- 5.Terabe M, Swann J, Ambrosino E, Sinha P, Takaku S, Hayakawa Y, Godfrey DI, Ostrand-Rosenberg S, Smyth MJ, Berzofsky JA (2005) A nonclassical non-Valpha14Jalpha18 CD1d-restricted (type II) NKT cell is sufficient for down-regulation of tumor immunosurveillance. J Exp Med 202(12):1627–1633. https://doi.org/10.1084/jem.20051381 Google Scholar
- 6.Matsuda JL, Naidenko OV, Gapin L, Nakayama T, Taniguchi M, Wang CR, Koezuka Y, Kronenberg M (2000) Tracking the response of natural killer T cells to a glycolipid antigen using CD1d tetramers. J Exp Med 192(5):741–754Google Scholar
- 8.Kawano T, Cui J, Koezuka Y, Toura I, Kaneko Y, Motoki K, Ueno H, Nakagawa R, Sato H, Kondo E, Koseki H, Taniguchi M (1997) CD1d-restricted and TCR-mediated activation of valpha14 NKT cells by glycosylceramides. Science 278(5343):1626–1629Google Scholar
- 9.Hayakawa Y, Godfrey DI, Smyth MJ (2004) Alpha-galactosylceramide: potential immunomodulatory activity and future application. Curr Med Chem 11(2):241–252Google Scholar
- 10.Smyth MJ, Crowe NY, Hayakawa Y, Takeda K, Yagita H, Godfrey DI (2002) NKT cells - conductors of tumor immunity? Curr Opin Immunol 14(2):165–171Google Scholar
- 11.Hayakawa Y, Takeda K, Yagita H, Kakuta S, Iwakura Y, Van Kaer L, Saiki I, Okumura K (2001) Critical contribution of IFN-gamma and NK cells, but not perforin-mediated cytotoxicity, to anti-metastatic effect of alpha-galactosylceramide. Eur J Immunol 31(6):1720–1727Google Scholar
- 12.Smyth MJ, Crowe NY, Pellicci DG, Kyparissoudis K, Kelly JM, Takeda K, Yagita H, Godfrey DI (2002) Sequential production of interferon-gamma by NK1.1(+) T cells and natural killer cells is essential for the antimetastatic effect of alpha-galactosylceramide. Blood 99(4):1259–1266Google Scholar
- 16.Maeda M, Shadeo A, MacFadyen AM, Takei F (2004) CD1d-independent NKT cells in beta 2-microglobulin-deficient mice have hybrid phenotype and function of NK and T cells. J Immunol 172(10):6115–6122Google Scholar
- 19.Vesely MD, Kershaw MH, Schreiber RD, Smyth MJ (2011) Natural innate and adaptive immunity to cancer. Annu Rev Immunol 29:235–271. https://doi.org/10.1146/annurev-immunol-031210-101324 Google Scholar
- 22.Blum JS, Wearsch PA, Cresswell P (2013) Pathways of antigen processing. Annu Rev Immunol 31:443–473. https://doi.org/10.1146/annurev-immunol-032712-095910 Google Scholar
- 27.Dunn GP, Old LJ, Schreiber RD (2004) The three Es of cancer immunoediting. Annu Rev Immunol 22:329–360. https://doi.org/10.1146/annurev.immunol.22.012703.104803 Google Scholar
- 29.Han RX, Liu X, Pan P, Jia YJ, Yu JC (2014) Effectiveness and safety of chemotherapy combined with dendritic cells co-cultured with cytokine-induced killer cells in the treatment of advanced non-small-cell lung cancer: a systematic review and meta-analysis. PLoS One 9(9):e108958. https://doi.org/10.1371/journal.pone.0108958 Google Scholar
- 30.Rosenberg SA, Lotze MT, Yang JC, Aebersold PM, Linehan WM, Seipp CA, White DE (1989) Experience with the use of high-dose interleukin-2 in the treatment of 652 cancer patients. Ann Surg 210(4):474–484 (discussion 484–475) Google Scholar
- 31.Rosenberg SA (1988) The development of new immunotherapies for the treatment of cancer using interleukin-2. A review. Ann Surg 208(2):121–135Google Scholar
- 36.Dominguez GA, Condamine T, Mony S, Hashimoto A, Wang F, Liu Q, Forero A, Bendell J, Witt R, Hockstein N, Kumar P, Gabrilovich DI (2017) Selective targeting of myeloid-derived suppressor cells in cancer patients using DS-8273a, an agonistic TRAIL-R2 antibody. Clin Cancer Res 23(12):2942–2950. https://doi.org/10.1158/1078-0432.CCR-16-1784 Google Scholar
- 42.Molon B, Ugel S, Del Pozzo F, Soldani C, Zilio S, Avella D, De Palma A, Mauri P, Monegal A, Rescigno M, Savino B, Colombo P, Jonjic N, Pecanic S, Lazzarato L, Fruttero R, Gasco A, Bronte V, Viola A (2011) Chemokine nitration prevents intratumoral infiltration of antigen-specific T cells. J Exp Med 208(10):1949–1962. https://doi.org/10.1084/jem.20101956 Google Scholar
Open AccessThis article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made.