Decitabine immunosensitizes human gliomas to NY-ESO-1 specific T lymphocyte targeting through the Fas/Fas ligand pathway

Abstract

Background: The lack of effective treatments for gliomas makes them a significant health problem and highlights the need for the development of novel and innovative treatment approaches. Immunotherapy is an appealing strategy because of the potential ability for immune cells to traffic to and destroy infiltrating tumor cells. However, the absence of well-characterized, highly immunogenic tumor-rejection antigens (TRA) in gliomas has limited the implementation of targeted immune-based therapies.

Methods: We hypothesized that treatment with the demethylating agent, decitabine, would upregulate the expression of TRA on tumor cells, thereby facilitating enhanced surveillance by TRA-specific T cells.

Results and discussion: Treatment of human glioma cells with decitabine increased the expression of NY-ESO-1 and other well characterized cancer testes antigens. The upregulation of NY-ESO-1 made these tumors susceptible to NY-ESO-1-specific T-cell recognition and lysis. Interestingly, decitabine treatment of T98 glioma cells also sensitized them to Fas-dependent apoptosis with an agonistic antibody, while a Fas blocking antibody could largely prevent the enhanced functional recognition by NY-ESO-1 specific T cells. Thus, decitabine treatment transformed a non-immunogenic glioma cell into an immunogenic target that was efficiently recognized by NY-ESO-1--specific T cells.

Conclusions: Such data supports the hypothesis that agents which alter epigenetic cellular processes may "immunosensitize" tumor cells to tumor-specific T cell-mediated lysis.

Figure 1

Treatment of human glioma with decitabine up-regulates cancer-testes antigens and MHC I. A) Total RNA was isolated from U-251MG glioma cells treated with decitabine or vehicle control and subjected to global gene expression classification using Affymetrix human U133 Plus 2.0 microarray chips. B) Decitabine-treated and untreated U-251MG glioma cells were stained with α-HLA-A, B, C and α-HLA-A2. Representative flow cytometric analysis data are shown here. Similar results were seen in two independent experiments.

Figure 2

Up-regulated expression of NY-ESO-1 in human glioma-derived cell lines and primary cell lines treated with decitabine. A) Representative data for established glioma cell lines; B) primary glioblastoma cells obtained from surgical resection; or C) normal human astrocytes (NHA) and a NY-ESO-1⁺ human melanoma line, 624.38. Only decitabine-treated glioma cells demonstrated expression of NY-ESO-1. D) Using SYBR green DNA labeling and quantitative PCR primers specific for NYESO-1 and GAPDH, the cDNA was amplified using a real-time PCR protocol. The relative fold change in gene expression is graphed showing an increase average fold change in treated T98G and glioblastoma #1 cells compared to untreated (***p = 0.0001 and **p = 0.007). Similar results were seen in three replicate experiments.

Figure 3

Expression of NY-ESO-1 in human brain tumor tissues. A large panel of normal tissues and brain tumors of varying grades were subjected to global gene expression profiling using Affymetrix U133 2.0 chips. The relative expression of NY-ESO-1 was normalized and tested. Dots plotted underneath the black line are considered to have background relative expression levels.

Figure 4

Retroviral transduction of PBMC with a cloned NY-ESO-1 specific T cell receptor allows for T cell recognition of decitabine-treated human T98 glioma cells. Normal donor peripheral blood mononuclear cells (PBMCs) were activated with OKT-3 at a concentration of 50 ng/mL for two days and then harvested for transduction. A) PBMCs were transduced with retroviral constructs encoding a HLA-A*0201-restricted NY-ESO-1 T cell receptor, and stained with fluorescent mAbs and tetramer NY-ESO-1 percentage. B) Representative flow cytometric analysis of CD107A of decitabine-treated and untreated T98G glioma cells (solid filled, NY-ESO-1 specific T cells alone; dashed line, NY-ESO-1 specific T cells co-cultured with untreated T98G glioma cells; solid line, NY-ESO-1 specific T cells co-cultured with decitabine-treated T98G glioma cells). Golgi-plug was added to co-cultures and stained with CD107A before incubation. The lymphocytes were stained with antibodies to CD3, CD4, and CD8 and subsequently fixed. The experiment was repeated three times with similar findings.

Figure 5

Elevated cytokine secretion of NY-ESO-1 TCR-transduced PBMCs co-cultured with decitabine-treated T98G glioma cells. T98 glioma cells were treated with or without decitabine and then co-cultured with NY-ESO-1 TCR-transduced T cells. Cell-free supernatants were tested for cytokine secretion by IFN-γ ELISA and cytometric bead array. A) Data from the IFN-γ ELISA is depicted (p <0.05). B) Concentrations of Th1/Th2 cytokines IFN-γ, TNF-α, and IL-5 (pg/mL) in co-cultured samples are depicted (***p < 0.001). Results shown are representative findings from over 5 replicate experiments.

Figure 6

Decitabine sensitizes glioma cells to an immune-mediated cell death via the Fas/FasL pathway. A) Decitabine-treated and untreated T98G cells were stained with an anti-Fas antibody and analyzed by flow cytometric analysis. B) T98G cells were treated as described previously and stained with an agonistic-Fas antibody (clone CH-11) to stimulate apoptosis for 24 hr. Results depict the average percentage of late cell death in untreated alone, decitabine-treated alone, untreated with agonistic mAb, and decitabine-treated with agonistic mAb. Decitabine-treated alone versus decitabine-treated with antibody (P = 0.0049). Untreated alone versus untreated with antibody (P = 0.0215). Experiments were conducted in triplicate. C) Representative flow cytometric analysis of CD107A staining from NY-ESO-1 TCR transduced T cells co-cultured with untreated T98G glioma cells, with or without the addition of an antagonistic-Fas antibody (clone ZB4) to block apoptosis (solid filled, NY-ESO-1 specific T cells alone; solid line, NY-ESO-1 specific T cells co-cultured with untreated T98G glioma cells; dashed line, NY-ESO-1 specific T cells co-cultured with untreated T98G glioma cells with antagonistic-Fas antibody. NY-ESO-1 TCR transduced T cells were gated from the CD8 population. D) Representative flow cytometric analysis of CD107A staining from NY-ESO-1 transduced T cells co-cultured with decitabine-treated T98 glioma cells, with or without the addition of an antagonistic-Fas antibody (clone ZB4) to block apoptosis (solid filled, NY-ESO-1 specific T cells alone; solid line, NY-ESO-1 specific T cells co-cultured with decitabine-treated T98G glioma cells; dashed line, NY-ESO-1 specific T cells co-cultured with decitabine-treated T98G glioma cells with antagonistic-Fas antibody. E) CBA was performed on cell-free supernatant of samples from Figure 6C. Th1/Th2 cytokine concentrations IFN-γ, TNF-α, and IL-5 (pg/mL) are depicted. (*p < 0.05; ** p < 0.01; NS, not significant).