Topotecan enhances immune clearance of gliomas

Abstract

Despite aggressive surgery, radiation therapy, and chemotherapy, glioblastoma multiforme (GBM) is refractory to therapy, recurs quickly, and results in a median survival time of only 14 months. The modulation of the apoptotic receptor Fas with cytotoxic agents could potentiate the response to therapy. However, Fas ligand (FasL) is not expressed in the brain and therefore this Fas-inducing cell death mechanism cannot be utilized. Vaccination of patients with gliomas has shown promising responses. In animal studies, brain tumors of vaccinated mice were infiltrated with activated T cells. Since activated immune cells express FasL, we hypothesized that combination of immunotherapy with chemotherapy can activate Fas signaling, which could be responsible for a synergistic or additive effect of the combination. When we treated the human glioma cell line U-87 and GBM tumor cells isolated from patients with TPT, Fas was up regulated. Subsequent administration of soluble Fas ligand (sFasL) to treated cells significantly increased their cell death indicating that these Fas receptors were functional. Similar effect was observed when CD3(+) T cells were used as a source of the FasL, indicating that the up regulated Fas expression on glioma cells increases their susceptibility to cytotoxic T cell killing. This additive effect was not observed when glioma cells were pre-treated with temozolomide, which was unable to increase Fas expression in tumor. Inhibition of FasL activity with the antagonistic antibody Nok-1 mitigated these effects confirming that these responses were specifically mediated by the Fas-FasL interaction. Furthermore, the CD3(+) T cells co-cultured with topotecan treated U-87 and autologous GBM tumor cells showed a significant increase in expression in IFN-gamma, a key cytokine produced by activated T cells, and accordingly enhanced tumor cytotoxicity. Based on our data we conclude that drugs, such as topotecan, which cause up regulation of Fas on glioma cells can be potentially exploited with immunotherapy to enhance immune clearance of tumors via Fas signaling.

Fig. 1

a Increased number (top number) of U-87 cells and mean fluorescent intensity (MFI) (bottom number) of Fas expression compared to the control of untreated cells. U-87 cells were treated with TPT (0.1 and 1.0 μM) for 24 h and then labeled with an anti-Fas antibody (Fas PE—light grey histogram for baseline and dark grey histogram for TPT treated U-87 cells) showing a shift in response to 0.1 and 1.0 μM TPT compared to the corresponding isotype control (white histogram). b Increased number of GBM cells expressing Fas compared to the control of untreated GBM. GBM cells were treated with TPT (0.1, 1.0 and 10 μM) for 24 h and then labeled with anti-Fas antibody showing a shift in the expression of FAS in response to TPT treatment and the corresponding isotype control. This experiment was replicated in three separate GBM patients with similar results

Fig. 2

a Soluble FasL (sFasL) enhances glioma cytotoxicity. U-87 cells were treated with either TPT (0.1, 0.3, 0.5 μM) or TMZ (100, 150, 200 μM) and sFasL was added 24 h later. The percentage of U-87 cell death was calculated using the microtetrazolium assay and compared to cells not treated with sFasL. The control group treated only with sFasL had <5% glioma cytotoxicity. Bars, SD; *P < 0.05 compared to untreated controls. b Cell death in U-87 glioma cells in response to TPT or TMZ in the presence or absence of stimulated or unstimulated CD3⁺ T cells. U-87 cells were pretreated with the appropriate chemotherapeutic agent and cultured with CD3⁺ T cells (stimulated or unstimulated) at 3:1 effector to target ratios. Glioma cells were labeled with CFSE and PI. CFSE+ glioma cells were gated and cell death was characterized by the percentage of CFSE⁺PI⁺ double positive cells. Tumor cells were treated with 0.1 μM of TPT in the presence or absence of stimulated or unstimulated CD3⁺ T cells. Bars, SD; *P < 0.05 compared to TPT-treated glioma cells (grey bars). There was significant difference (P < 0.05) between the percentage of cell death with stimulated and unstimulated T cells in U-87 cells pretreated with TPT. Tumor cells were treated with 150 μM of TMZ in the presence or absence of unstimulated or stimulated CD3⁺ T cells. Bars, SD; there were no statistically significant differences when either unstimulated or T cells or stimulated T cells were added

Fig. 3

The CD3⁺ T cells were co-cultured with a FasL blocking agent, Nok-1 and TPT-treated U-87 cells were exposed to unstimulated or stimulated CD3⁺ T cells. Bars, SD; *P < 0.05 (compared to TPT treated glioma cells with stimulated CD3⁺ T cells). All experiments were replicated in their entirety three times

Fig. 4

IFN-γ production by CD3⁺ T cells after incubation with TMZ or TPT pretreated glioma cells. CD3⁺ T cells from normal donors were incubated with U-87 cells pretreated with the respective chemotherapeutic agent. CD3⁺ T cells were gated and CD3⁺IFN-γ⁺ T cells (shaded histogram) were compared to the respective isotype control (clear histogram) by flow cytometry analysis. All experiments were replicated in their entirety three times will identical results

Fig. 5

Representative proliferation and cytokine production in T cells from glioblastoma patients and normal donors. a CFSE-labeled CD8⁺ T cells isolated from normal donor PBMCs were incubated with unstimulated autologous APCs (CD11b⁺), in the presence of anti-CD3 antibody or in the presence of anti-CD3 antibody with the co-stimulation anti-CD28/CD49d antibodies (top panel) CFSE-labeled CD8⁺ T cells isolated from newly diagnosed GBM patients were incubated under similar conditions (bottom panel). Cells were surface stained for T cell markers (CD3, CD8) and gated CD8⁺ T cells were analyzed for cell proliferation by flow cytometry via CFSE dilution. Numbers on the plots indicate the percentage of gated cells that have undergone cell division. These data are representative of three separate experiments. b GBM patients or normal donors PBMCs were isolated and stimulated with antigen (tetanus toxoid) and then stained for CD8 and intracellular IFN-γ expression. Using flow cytometry analyses, the percentage of CD8⁺ IFN-γ⁺ double positive cells were identified in the top right quadrant of each plot. There was significant difference (P < 0.05) between CD8⁺ IFN-γ⁺ responding T cells from GBM patients and normal donors IFN-γ production by CD3⁺ T cells. c Flow cytometry panels showing IFN-γ production by PBMCs from a representative GBM patient after incubation with their respective TPT-pretreated GBM at a tumor to PBMC ratio of 1:5. An increase in intracellular IFN-γ was determined by flow cytometry analysis. Similar data was obtained in three separate GBM patients

Fig. 6

Cell death in glioma cells pretreated with TPT when cultured with CD3⁺ T cells from patients with glioblastoma. a U-87 cells were pretreated with TPT at 0.1 μM were then cultured at 3:1 effector to target ratios with CD3⁺ T cells isolated from PBMCs of newly diagnosed GBM patients. This experiment was replicated three times with similar findings. b GBM tumors were placed into a single cell suspension and pretreated with TPT at 1.0 μM and then cultured at 5:1 effector to target ratios with the patient’s own CD3+ T cells. This experiment was replicated twice and a representative specimen is shown. Bars, SD; *P < 0.05 compared to TPT treated glioma cells