Ionizing radiations sustain glioblastoma cell dedifferentiation to a stem-like phenotype through survivin: possible involvement in radioresistance

Abstract

Glioblastomas (GBM) are some bad prognosis brain tumors despite a conventional treatment associating surgical resection and subsequent radio-chemotherapy. Among these heterogeneous tumors, a subpopulation of chemo- and radioresistant GBM stem-like cells appears to be involved in the systematic GBM recurrence. Moreover, recent studies showed that differentiated tumor cells may have the ability to dedifferentiate and acquire a stem-like phenotype, a phenomenon also called plasticity, in response to microenvironment stresses such as hypoxia. We hypothesized that GBM cells could be subjected to a similar dedifferentiation process after ionizing radiations (IRs), then supporting the GBM rapid recurrence after radiotherapy. In the present study we demonstrated that subtoxic IR exposure of differentiated GBM cells isolated from patient resections potentiated the long-term reacquisition of stem-associated properties such as the ability to generate primary and secondary neurospheres, the expression of stemness markers and an increased tumorigenicity. We also identified during this process an upregulation of the anti-apoptotic protein survivin and we showed that its specific downregulation led to the blockade of the IR-induced plasticity. Altogether, these results demonstrated that irradiation could regulate GBM cell dedifferentiation via a survivin-dependent pathway. Targeting the mechanisms associated with IR-induced plasticity will likely contribute to the development of some innovating pharmacological strategies for an improved radiosensitization of these aggressive brain cancers.

Figure 1

Characterization of the stem and differentiated phenotypes in GSC-enriched NS and FCS-differentiated GBM cultures. (a–c) GSC-enriched NS cell lines isolated from four patient tumors (C, D, G and I) were kept in SCM medium or allowed to differentiate as adherent GBM cells for at least 15 days in FCS medium. (a) Phase-contrast photomicrographs of NS or GBM-differentiated cells. Original magnification: × 10, scale bar: 6 μm. (b) Real-time quantitative PCR analysis of the stem (CD133, Notch1, Nanog, Gli1, Sox2, Nestin, SonicHH, EZH2, Olig1 and Olig2) and differentiation (GFAP and CTGF) markers in NS or GBM-differentiated cells for the C, D, G and I cell lines. Shown are the fold inductions expressed as means±S.E.M. of at least three independent experiments. ^∗P<0.05, ^∗∗P<0.01, ^∗∗∗P<0.001 compared with the related control. (c) Immunofluorescence FACS analysis of stem (CD133, Notch1, Nanog, Sox2, Nestin and A2B5) and differentiation (GFAP, Tuj1 and O4) markers in NS or GBM-differentiated cells. The results depicted were representative of three independent experiments (G cell line) and were reproduced in all the cell lines

Figure 2

Overview of the long-term dedifferentiation protocol and assessment of the irradiation dose. (a) As described in the Materials and Methods section, GSC-enriched NS were isolated from patient samples and cultured in a specific SCM medium. NS cells were then dissociated and placed into a differentiating medium with FCS (FCS medium) for at least 15 days, to allow an optimum differentiation. Adherent differentiated GBM cells were then subjected or not to a 3-Gy irradiation and were placed 2 days after in fresh FCS medium, to keep them fully differentiated, or in SCM medium to favor a possible dedifferentiation process. The totality of the cells was finally collected at the end of the protocol, which coincided to the apparition of NS in the culture supernatant, in order to be analyzed. (b–f) Absence of effects of a 3-Gy irradiation on viability and proliferation of GBM cells during the dedifferentiation protocol. Differentiated GBM cells were subjected or not (Control, CTR) to a 3-Gy irradiation and placed 2 days after in SCM medium for 5 additional days, according to the dedifferentiation protocol (b–d) or for different time points during this dedifferentiation protocol (e and f) The dose of 12 Gy was chosen as a positive control for cell death induction. At 7 days post IR, cells were analyzed either by FACS for AV/PI double staining of both apoptotic and necrotic cells (b) and PI staining of the Sub-G1 population (c), or through WST-1 staining using a spectrophotometric determination of the dye absorbance at 450 nm for the quantification of cell viability and proliferation (d). For kinetic studies, cell death was assessed by FACS, by AV/PI staining and expressed as percentages over background (e), and cell count was performed to estimate the cell proliferation rate in the indicated culture condition (f). Shown are the means±S.E.M. of at least three independent experiments. ^∗P<0.05, ^∗∗P<0.01 compared with the related control

Figure 3

Increased ability to generate NS in GBM-differentiated cells subjected to a 3-Gy subtoxic irradiation. Cells treated or not by a 3-Gy irradiation and placed 2 days after in either FCS or SCM medium for long-term culture were analyzed in order to evaluate the number of NS generated in the culture supernatant. (a) Phase-contrast photomicrographs of GBM cells at the end of the dedifferentiation protocol. Arrows indicate the presence of NS. Original magnification: × 10, scale bar: 6 μm. (b–d) Quantification of the number of generated NS at the end of the dedifferentiation protocol for the four cell lines C, D, G and I in irradiated (3 Gy) or untreated (CTR) GBM cells kept in SCM medium. (b) NS were counted and results are expressed per 25-cm² flasks. Results are expressed as the means±S.E.M. of at least three independent experiments. ^∗P<0.05, ^∗∗P<0.01 compared with the related control. (c) NS were also generated from differentiated cells through the use of a dilution assay at low density and were counted at different dilutions in each well of a 24-well plate. The results are shown for the C and I cell lines (20 000 cells/well). ^∗P<0.05. (d) Before the NS generation assay, A2B5-negative differentiated cells were sorted as described in the Materials and Methods section (A cell line). The generated primary NS were then counted at the end of the dedifferentiation protocol. Results are expressed as the means±S.E.M. of three independent experiments. ^∗P<0.05 (e and f) Primary NS were generated from differentiated cells with (f) or without (e) an optional FACS sorting of the A2B5-negative differentiated cells. These primary NS were subsequently dissociated and plated in 96-well plates at different low cell densities, to study their ability to generate secondary NS through limiting dilution assays. ^∗P<0.05. ^∗∗P<0.01, ^∗∗∗P<0.001 compared with the related CTR SCM condition. Representative phase-contrast photomicrographs were shown for each conditions (original magnification: × 4, scale bar: 17 μm)

Figure 4

Overexpression of stemness markers and downregulation of differentiation markers at the RNA level in GBM cells after a 3-Gy irradiation. Differentiated GBM cells treated or not by a 3-Gy irradiation and placed 2 days after in either FCS or SCM medium for long-term culture were analyzed by real-time quantitative PCR (see Materials and Methods) at the end of the dedifferentiation protocol. RNA expression level of the indicated stem (a) or differentiation markers (b) in GBM-differentiated cells subjected to the dedifferentiation process for the indicated cell lines. The RNA expression levels of these different markers were also shown for NS cells as a control, as these NS are enriched in GSC. Shown are the fold inductions relative to the CTR SCM condition expressed as means±S.E.M. of at least three independent experiments. ^∗P<0.05, ^∗∗P<0.01, ^∗∗∗P<0.001 compared with the related CTR SCM condition

Figure 5

Overexpression of stemness markers and downregulation of differentiation marker at the protein level in GBM cells after a 3-Gy irradiation. Differentiated GBM cells treated or not by a 3-Gy irradiation and placed 2 days after in either FCS or SCM medium for long-term culture were analyzed either by western blotting (a) or FACS immunofluorescence (b) at the end of the dedifferentiation protocol. Protein expression levels in NS cells were shown as a control for the stem condition. (a) Western blotting analysis of the stem markers Olig2, Sox2 and Nestin. Equal gel loading and transfer efficiency were checked with anti-actin or β2-microglobulin (β2M) antibodies. Blots were representative of at least three independent experiments in the indicated cell line and were reproduced in all the cell lines. (b) Immunofluorescence analysis performed by FACS of the stem (Nestin, A2B5, Nanog and Notch1) and differentiation (GFAP) markers in the G cell line. The SFI allowed to evaluate the marker expression level (see Materials and Methods). Results are expressed as the means±S.E.M. of at least three independent experiments in the G cell line and were reproduced in all the cell lines. ^∗P<0.05, ^∗∗P<0.01 compared with the related CTR SCM condition. For each marker, some representative FACS plot overlays (NS versus CTR FCS, CTR FCS versus 3 Gy FCS and CTR SCM versus 3 Gy SCM) were depicted

Figure 6

Increased in vivo tumorigenicity of 3-Gy-irradiated GBM cells placed in SCM for long-term culture. Differentiated GBM cells treated or not by a 3-Gy irradiation and placed 2 days after in either FCS or SCM medium for long-term culture were subsequently orthotopically xenografted in nude mice to evaluate their tumorigenic potential. NS cells were also injected as a control for the stem condition, as they are enriched in GSC. (a) Survival curves established in xenografted mice for the indicated injected cell line (three mice per group for the D cell line and five mice per group for the G cell line). Exact P-values between the 3-Gy SCM group and the related CTR SCM group are indicated in the figure, after log-rank analysis. (b) Immunohistochemistry (IHC) analysis of Nanog-positive cell clusters in the brain tumors of killed, xenografted mice for the CTR SCM and the 3-Gy SCM groups (D and G cell lines, three mice per group). Shown are means±S.E.M. ^∗P<0.05

Figure 7

Overexpression of the anti-apoptotic protein survivin in 3-Gy-irradiated GBM-differentiated cells placed in SCM for long-term culture. Differentiated GBM cells treated or not by a 3-Gy irradiation and placed 2 days after in either FCS or SCM medium for long-term culture were analyzed for survivin expression either by real-time quantitative PCR (a) or western blotting (b) at the end of the dedifferentiation protocol and for the four different patient cell lines. RNA and protein expression levels for Survivin were also analyzed in NS cells as a control for the stem condition. (a) PCR results were expressed as fold inductions relative to the CTR SCM condition and shown as means±S.E.M. of at least three independent experiments. ^∗P<0.05, ^∗∗P<0.01, ^∗∗∗P<0.001. (b) Western blotting results were representative of at least three independent experiments for each cell line. Equal gel loading and transfer efficiency were checked with an anti-actin antibody

Figure 8

Requirement of the IR-induced survivin overexpression for the dedifferentiation process in GBM cells. (a–d) Differentiated GBM cells were pre-treated for 24 h with either a survivin inhibitor (YM-155 7 nM) or an AKT inhibitor (MK-2206 250 nM) and then irradiated or not at 3 Gy before being placed 2 days after in either FCS or SCM medium, complemented or not with fresh YM-155 or MK-2206 inhibitors. (a) Western blot analysis of the effect of YM-155 and MK-2206 treatment on Survivin expression in GBM cells (I cell line) irradiated or not and kept in SCM medium for 1 additional week. Efficiency of MK-2206 toward AKT was also checked as a control by the blotting of phospho-AKT1 (pAKT1). (b and c) At the end of the dedifferentiation protocol, the effects of YM-155 and MK-2206 were measured on the NS generation potential in response to IR by NS counting in phase-contrast microscopy (original magnification: × 4, scale bar: 17 μm) (b) and subsequent quantification in the indicated cell lines (c). Results are expressed as the means±S.E.M. of three independent experiments. ^∗∗P<0.01, ^∗∗∗P<0.001 compared with the 3-Gy SCM condition. (d) The involvement of Survivin in the IR-induced GBM reprogramming was checked by western blotting by analyzing the expression of the stem markers Nestin, Sox2 and Olig2 at the end of the dedifferentiation protocol in the presence or absence of YM-155 and MK-2206. Concerning western blottings, equal gel loading and transfer efficiency were checked with an anti-actin, AKT1 or β2-microglobulin (β2M) antibody, and results were representative of at least three independent experiments on the indicated cell line and reproduced in all the cell lines