W.Z. and B.L. contributed equally to this work
Induction of autophagy promotes differentiation of glioma-initiating cells and their radiosensitivity
Article first published online: 8 APR 2011
Copyright © 2011 UICC
International Journal of Cancer
Volume 129, Issue 11, pages 2720–2731, 1 December 2011
How to Cite
Zhuang, W., Li, B., Long, L., Chen, L., Huang, Q. and Liang, Z. (2011), Induction of autophagy promotes differentiation of glioma-initiating cells and their radiosensitivity. Int. J. Cancer, 129: 2720–2731. doi: 10.1002/ijc.25975
- Issue published online: 28 SEP 2011
- Article first published online: 8 APR 2011
- Accepted manuscript online: 3 FEB 2011 01:59PM EST
- Manuscript Accepted: 10 JAN 2011
- Manuscript Received: 20 SEP 2010
- National Natural Science Foundation of China. Grant Numbers: 30873052, 81072656
- Natural Science Foundation of Jiangsu Province, China. Grant Number: BK2010226
- Graduate Education Innovation Project of Jiangsu Province, China. Grant Number: CX10B_057Z
- Graduate Education Project of Soochow University. Grant Number: 23320935
- glioma-initiating cells;
Glioblastoma (GBM) is a highly aggressive brain tumor characterized by increased proliferation and resistance to chemotherapy and radiotherapy. Recently, the identification of tumor-initiating cells with stem-like properties in diverse human cancers including GBM represents an important conceptual advance in cancer biology with therapeutic implications. However, the factors determining the differential development and radiosensitization of glioma-initiating cells (GICs) remain poorly defined. Here, we report that rapamycin induced differentiation of GICs and increased their sensitivity to radiation by activating autophagy. Transient in vitro exposure to rapamycin and radiation abolished the capacity of transplanted GICs to establish intracerebral GBMs. Most importantly, in vivo combination of rapamycin and radiation effectively blocked the tumor growth and associated mortality that occurs in mice after intracerebral grafting of human GICs. We demonstrate that rapamycin activated their autophagy and triggers the differentiation cascade in GICs isolated from human GBMs. This was followed by a reduction in proliferation, cell viability, clonogenic ability and increased expression of neural differentiation markers after radiation. Our results suggest that autophagy plays an essential role in the regulation of self-renewal, differentiation, tumorigenic potential and radiosensitization of GICs, suggesting autophagy could be a promising therapeutic target in a subset of GBMs. We propose that autophagy defect in GICs contributes to radioresistance of GICs by desensitizing GICs to normal differentiation cues. Activating autophagy may abrogate the resistance of GICs to radiation and could lead to the development of novel therapeutic approaches for the treatment of GBMs.
Glioblastomas (GBMs) are common and aggressive brain tumors in adults. The current standard treatments for malignant gliomas include surgical resection, radiation therapy and chemotherapy. Despite progress in understanding of the molecular mechanisms involved in the genesis and progression of glioma, the prognosis and treatment of this tumor type continue to be dismal. The median survival of patients with gliomas is only 9–12 months.1 Recently, a growing body of literature indicates that only a small subpopulation of malignant glioma cells has true tumorigenic potential. This cell population, called glioma-initiating cells (GICs), brain tumor-initiating cells (TICs) or glioma stem cells, is considered to be responsible for the initiation, propagation and recurrence of tumors. GICs are characterized by self-renewal, multilineage differentiation, maintained proliferation and highly oncogenic potential.2, 3 GICs show resistance toward differentiation cues.4, 5 The diminished tumorigenicity of these TICs on restoration of differentiation potential, along with recent reports supporting prodifferentiation as a potential strategy to inhibit GBM-derived TICs,4–6 encourages the identification and testing of agents targeting these differentiation pathways in the treatment of GBM in humans.
Autophagy is a dynamic process of protein degradation, which is essential for survival, differentiation, development and homeostasis. Accumulating evidence indicates that alterations in autophagy may influence cell fate during mammalian development and differentiation. Autophagy genes could play a role in the specification of the left–right axis. The factor UVRAG (a component of the dynamic Beclin 1/Vps34 complex)7 is mutated in a human case of abnormal left–right axis formation, resulting in heterotaxy and multiple malformations.8 The first analysis of atg5 embryonic expression showed high levels of atg5 mRNA in the zone of polarizing activity, the area of the limb bud responsible for anteroposterior axis formation.9 A recent study has reported that embryonic stem cells lacking Beclin 1 or atg5 are defective in forming cavitated embryoid bodies in vitro because of the failure in clearing apoptotic cells.10 Furthermore, the Beclin 1 level confers both autophagy and differentiation capabilities in leukemia cells.11 These studies suggested that autophagy has an important role in development and differentiation, although its specific functions remained unclear. The differentiation pathways of GICs are complex and differentially regulated by many cytokines. Therefore, a puzzling question is whether autophagy plays a role in differentiation of GICs.
Recent studies suggest that autophagy is important in the regulation of cancer development and progression and in determining the response of tumor cells to anticancer therapy. Also, autophagy is a novel response of malignant glioma cells to ionizing radiation (IR). However, the outcome of autophagy observed in glioma cells after IR is not straightforward.12, 13 Although association between autophagy and radiosensitization was demonstrated, the precise role of autophagy in relation to glioma cell death is hard to define. A core molecule in autophagy regulation is the kinase mammalian target of rapamycin (mTOR). mTOR inactivation has been shown to initiate autophagy.14, 15 mTOR inhibitors such as rapamycin and rapamycin-derivative RAD001 (Everolimus) are known for their antiangiogenic effects16 that can sensitize cancer cells to radiation effects.17 Rapamycin is used in Phase II trials for glioma therapy.18 One group completed and published a clinical trial of the mTOR inhibitor rapamycin in patients. They showed that the mTOR inhibitor rapamycin was present in potentially therapeutic levels in tumor tissue in vivo.19 However, early clinical trials with the mTOR complex inhibitor rapamycin and its derivatives did not show impressive clinical responses.20, 21 This was not due to cell intrinsic resistance, but more likely was associated with failure of the drug to fully access its target in vivo.19
In our study, we examined the induction of autophagy by radiation and its role in the radioresistance of GICs. Contrary to the previous observations that γ-radiation induced a larger degree of autophagy in the CD133+ GICs as compared to CD133− cells,22 we found that GICs expressed lower levels of the autophagy-related proteins, LC3. Moreover, radiation induced a low degree of autophagy in these cells. By the induction of autophagy via mTOR inhibitor, rapamycin could trigger differentiation of GICs and enhance their radiosensitization in vitro and immunocompromised mice model. We propose that autophagy defect in GICs contributes to the differentiation block and glioma radioresistance. To our knowledge, this is the first evidence showing that upregulation of autophagy is required in differentiation cascades of GICs. Our results suggested that the delivery of rapamycin combined with radiotherapy is a potential therapeutic strategy to enhance current therapy in patients with GBM.
Material and Methods
Tumor samples classified as GBM based on the World Health Organization criteria were obtained from patients undergoing surgical treatment. The subjects' consent was obtained according to the Declaration of Helsinki. The study protocol was approved by the Medical Review Board of Soochow University Medical School.
GBM neurospheres were generated as described previously.23 Cells proliferated in the presence of fibroblast growth factor (FGF, 20 ng/ml; Gibco Life Technologies, Paisley, UK), epidermal growth factor (EGF, 20 ng/ml; Invitrogen, Camarillo, CA) and N2 supplement (Gibco Life Technologies, Paisley, UK) in serum-free media Dulbecco's Modified Eagle Media (DMEM)/F12 (Gibco) and differentiated in the presence of serum or in the absence of bFGF and EGF.
Antibodies and reagents
The following antibodies were used: nestin (Millipore, Temecula, CA), (clone 10C2), Musashi (EP1302) (Millipore, Temecula, CA), O4 (Millipore, Temecula, CA), (clone 81), GFAP, (Therom, Rockford, USA) (clone GFA-02), MAP-2 (HM-2) (Sigma, St. Louis, MO), p-H2AX (Millipore, Temecula, CA), (clone JBW301) (Therom, Rockford, USA), β-tubulin (clone 2G10) (Millipore, Temecula, CA) and β-actin (clone EP1123Y) (Millipore, Temecula, CA). Rapamycin (Cat: R0395) (Sigma, St. Louis, MO) pretreatment was performed by adding 200 nM of rapamycin to growth medium 48 hr before transplantation. Rapamycin was added from 1 mg/ml stock in Dimethyl sulfoxide (DMSO) Sigma, St. Louis, MO. Final concentration of DMSO in the growth medium was 0.01%.
Radiation treatment, clonogenic survival and cell viability assay
Cells or mice were irradiated with a 6-MV X-ray linear accelerator (model: PRIMUS, D. E., Siemens A&D LD) at dose rate of 198 MV/min. Clonogenic survival assay and cell viability assay were performed as described previously24 (see Supporting Information).
Western blot analysis
Immunoblotting was done as described previously.25
Tissue blocks were cut serially into ultrathin (0.06 μm) sections and then analyzed by a Philips CM-120 electron microscopy as described previously.25
Annexin V-FITC/PI staining
The percentage of apoptosis was measured using the Annexin fluoresceine isothiocyanate V-FITC/PI (propidium iodide) apoptosis detection kit (Vybrant® Apoptosis Assay Kit #2; San Diego, USA. BD Biosciences) with flow cytometry. Five thousand cells per sample were analyzed using a flow cytometer (Epics-XL, Beckman).
All quantitative data presented are the mean ± SD from at least three samples per data point. The difference was considered statistically significant when p < 0.05. Survival data were entered into Kaplan–Meier plots, and statistical analysis was performed using a log-rank test. A p value of 0.05 was used as the boundary of statistical significance.
Radiosensitization of GICs by rapamycin and radiation
We obtained GICs, named SU-2 and SU-3, from surgically resected human GBM samples as reported previously.23 Cells cultured in the EGF-, FGF- and N2-supplemented serum-free medium readily generated nonadherent, multicellular spheres (neurospheres; Supporting Information Fig. S1A). Neurospheres originating from the tumor specimens expressed high levels of the cancer stem/progenitor cell markers2, 26–28 CD133, nestin, Sox2 and Musashi (Supporting Information Fig. S1B) and underwent multilineage differentiation acquiring the expression of O4 (oligodendrocytic marker), GFAP (astrocytic marker) and Map-2 (neuronal marker) when cultured in the presence of serum (Supporting Information Fig. S1C).
To determine if the combination of autophagy induction and IR leads to radiosensitization in GICs, clonogenic survival assay was performed. We examined the effects of radiation, rapamycin, rapamycin + IR and rapamycin + IR + 3-MA on cell survival. To demonstrate this clonogenic assay studies, a Student's t-test was performed on the Surviving Fraction (SF) values. The analysis was performed comparing SF (untreated, SF = 1) with SF (radiation, SF = 0.81), SF (rapamycin, SF = 0.83) or SF (rapamycin + IR + 3-MA, SF = 0.82) at either 6 Gy; in each case, the difference was not statistically significant (p > 0.05). Cells previously treated with rapamycin and IR generated a decreased number of colonies as compared to untreated cells (p < 0.05, SF = 0.09 for rapamycin + IR). This statistical analysis supports the conclusion derived by comparison of the radiation dose enhancement ratio (DER) values. The DER in the rapamycin + IR + 3-MA (3-methyladenine) treatment group compared to radiation alone was 0.99. Administration of both radiation and rapamycin resulted in a DER of 9.0 (p < 0.005, Student's t-test, n = 3) compared to radiation alone treatment group (Fig. 1a). To assess the killing effect of 6-Gy radiation in the short term, we did a cell viability assay up to 3 days after the treatments. The combination (rapamycin + IR) inhibited cell viability more than each treatment did alone. The viability of GICs treated with rapamycin and radiation decreased to 40% compared to 89% of viability in radiation alone group, 84% of viability in rapamycin alone group and 80% of viability in triple treatment group (rapamycin + IR + 3-MA) after 3 days (Fig. 1b). These findings suggested that rapamycin leads to increased radiation sensitivity in GICs.
Increased autophagy induction in GICs and in xenograft tumors following the administration of rapamycin and radiation
To distinguish specific mTOR-mediated cell proliferation block from nonspecific toxicity of rapamycin, we measured the phosphorylation of one of the downstream targets of mTOR, p70S6K, using Western blotting for GICs treated with rapamycin at 2–200 nmol/L for 24 hr. As shown in Supporting Information Figure S2, the phosphorylation of p70S6K at Thr389 in GICs was inhibited in a dose-dependent manner. In GICs, treatment with 100 nmol/l rapamycin remarkably suppressed the phosphorylation of p70S6K. We decided to treat cells with rapamycin at concentrations of 200 nM for subsequent experiments.
Increasing evidence indicates that inhibition of mTOR is associated with the induction of autophagy.29–32 We therefore investigated whether treating GICs with rapamycin and radiation would result in autophagy. Although microtubule-associated protein light chain 3 (LC3) has several homologs in mammals, LC3-II is most commonly used for autophagy assays. LC3-II, or the protein tagged at its N terminus with a fluorescent protein, has been used to monitor autophagy through indirect immunofluorescence. Another approach is to detect LC3 conversion (LC3-I to LC3-II) by immunoblot analysis because the amount of LC3-II is clearly correlated with the number of autophagosomes.33 After exposure of GICs to untreated, IR, rapamycin, IR + rapamycin, IR + rapamycin + 3-MA or IR + rapamycin + E64d, changes in cellular localization of punctate LC3 expression were observed (Fig. 2a). Quantification of the percentage of punctate LC3 expression in each treatments (Fig. 2b) showed that untreated and single treatment with radiation resulted in approximately 4 and 5% of punctuate LC3 expression, respectively, compared to 70 and 78% of punctuate LC3 expression in single treatment with rapamycin and cotreatment with radiation and rapamycin in SU-2. Similar results were observed in SU-3. To determine whether elevation in LC3 expression resulted from autophagy induction or sluggish autophagy flux, we used 3-MA to inhibit autophagy induction and used E64d, lysosomal protease inhibitors, to inhibit the autophagy flux. As expected, the expression of LC3 was blocked by 3-MA, which inhibits Type III PI3K, initiator of autophagy. LC3 expression was elevated by E64d, which blocked the flux of autophagy. The changes in protein levels of LC3-I (18 kDa) and LC3-II (16 kDa) in GICs with the same treatments as mentioned above were further confirmed by the immunoblotting analysis (Fig. 2c).
We next performed xenograft experiments to investigate the effect of combined treatment with rapamycin and IR in animal models. Electron microscope showed a remarkable increase in the number of autophagosomes in tumor sections obtained from our in vivo tumor study on combination treatment (Fig. 2d).
Increased autophagy decreases tumorigenicity of GICs
At this point, we decided to address whether the in vitro reduction in GICs would correspond to a similar decline in the ability of cotreated cells to initiate tumors in vivo. To test this, GICs previously left untreated, or treated with rapamycin, IR, rapamycin + IR or rapamycin + IR + 3-MA, were inoculated in the brains of nude mice as described in the Supporting Information. All animals receiving untreated GICs developed large tumor masses that showed characteristic GBM features, including nuclear atypia, expression of aberrant glia, extensive neovascularization, and invaded the lateral, third and fourth ventricles (Fig. 3a).26 The tumors in the rapamycin or IR treatment alone group and those received triple treatment (rapamycin + IR + 3-MA) did not show any histomorphologic alterations compared to untreated control (Fig. 3a). Conversely, cotreated cells (rapamycin + IR) did not form invasive tumors, but formed small delimited lesions that were confined to the injection site, and showed no ventricular invasion (Fig. 3a). Animal survival bearing GICs intracranial xenografts was evaluated using a log-rank analysis from a Kaplan–Meier survival curve (Fig. 3b). Mice inoculated with GICs pretreated with rapamycin and IR exhibited significantly increased survival compared to mice inoculated with untreated cells, IR-treated cells or rapamycin-treated cells. Thus, transient exposure to rapamycin and IR depleted the GICs population and produced a significant decrease in the in vivo tumor-initiating ability of GBM cells.
Increased survival of mice bearing GICs intracranial xenograft tumor treated with rapamycin and IR
We next sought to establish if delivery of rapamycin in combination with IR in vivo might prevent intracerebral glioma tumor establishment and growth. In all experiments, animals receiving untreated, rapamycin alone, IR alone or triple therapy (rapamycin + IR + 3-MA) developed large malignant tumors surrounded by numerous small-satellite tumors, whereas mice treated with double therapy (rapamycin + IR) displayed small confined lesions (Supporting Information Fig. S3A). Lesions in cotreated animals included few neoplastic cells, conversely, tumors in the other groups contained pleiomorphic neoplastic elements with malignant infiltrating cells (Supporting Information Fig. S3A). Treatment of intracranial glioma bearing mice with untreated, rapamycin alone, IR alone or triple therapy (rapamycin + IR + 3-MA) leads to cachexia and poor feeding that does not improve, whereas double therapy (rapamycin + IR) results in less toxicity. Nearly all mice receiving cotreated cells survived, whereas 2–3 months postinjection, all animals in the other groups died (Supporting Information Fig. 3B). These data demonstrate that combination treatment of nude mice bearing intracranial GICs glioma xenografts with rapamycin and IR results in a significant increase in animal survival compared to treatment with rapamycin alone or IR alone.
Apoptosis detection assay
To test whether the increased response to radiation seen in vitro is associated with apoptosis, the level of apoptosis in GICs treated as described above were measured. Apoptosis level in GICs was significantly increased by the combined treatment with rapamycin and IR, as indicated by increased activation of caspase-3, an indicator of cell apoptosis (Fig. 4a). A corroborative assay, annexin V-conjugated FITC and propidium iodide staining was performed to measure the percentage of apoptosis in each treatment groups. IR therapy alone group, rapamycin alone group and triple therapy (rapamycin + IR + 3-MA) showed a slight increase in apoptosis level compared to untreated group, whereas apoptosis in cotreated cells was 20- to 30-fold greater than in matched untreated cells (Figs. 4b and 4c). These data indicate that radiation sensitization in cotreated GICs was due to higher rates of apoptosis.
GICs treated with rapamycin and IR repair DNA damage less efficiently than GICs in other groups
Although IR or drugs damaged tumor cells through several mechanisms, IR or drugs killed cancer cells primarily through DNA damage. Previous study has shown that CD133+ cells repaired the DNA damage more efficiently than CD133− cells.34 We compared the recovery of GICs in various groups in response to DNA damage using the alkaline comet assay.35 GICs in each group were equally susceptible to DNA damage initially, but the percentage of cells with comet tails decreased 4–5 times more rapidly in IR alone group, rapamycin alone group and triple therapy (rapamycin + IR + 3-MA) than in matched cotreated group (Fig. 5), indicating that cotreated GICs repaired the DNA damage less efficiently than GICs in other groups. This result was further confirmed by assessing the resolution of phosphorylated histone 2AX nuclear foci36 after various treatments (Supporting Information Fig. S4). The presence of phosphorylated H2AX nuclear foci in GICs in cotreated group was significantly higher than that in GICs in other groups at 24 hr after treatments. These data indicated the resolution of the phosphorylated H2AX nuclear foci in GICs cotreated with rapamycin and IR was much more slowly than that in the matched GICs in other groups. Decelerated resolution of the phosphorylated H2AX nuclear foci in GICs in cotreated group suggested less efficient repair in this population. The ability to repair DNA damage is essential to cellular survival because maintained DNA breaks induces apoptosis or senescence.37–39
Treatment with rapamycin and IR attenuated neurosphere formation capacity of GICs and induced differentiation of GICs
We decided to evaluate the effect of combined IR and rapamycin on the self-renewal of GICs following a well-described protocol based on the ability of GICs to generate neurospheres.40 Cells treated as described in the Supporting Information were dissociated into single cells and cultured in serum-free media with FGF2 and EGF. After 7 days, the newly formed neurospheres and the total number of cells were counted. Treatment with rapamycin and IR decreased the amount and size of the newly formed neurospheres as well as the total number of cells, indicating that the cotreatment represses self-renewal of GICs (Figs. 6a and 6b).
Next, we assessed the effect of various treatments on differentiation following the two abovementioned differentiation protocols. Neurospheres were incubated with FGF2 and EGF under serum-free condition. A time-course experiment showed that cells treated with rapamycin or treated with IR and rapamycin acquired the markers of differentiation Tuj1, GFAP and Olig2 (Fig. 6c) later than untreated cells and IR-treated cells. Relative to cells in untreated group and IR-treated group, cells in rapamycin-treated group and in cotreated group became more attached to the culture plate, losing the spherical shape (Fig. 6b) after 3 days of the indicated treatments. We also observed an increase in the expression of differentiation markers (GFAP and β-tubulin), together with a decrease in the expression of neural stem/progenitor markers (CD133 and Nestin) as detected by immunocytochemical assays (Fig. 6d). Overall, our results indicated that combined IR and rapamycin not only deregulated the self-renewal of GICs but also induced the differentiation of GICs. Thus, high autophagy levels enhanced the expression of markers of differentiation, corroborating their effect on GICs self-renewal.
Treatment with IR and rapamycin decreases tumorigenicity of GICs through induction of differentiation of GICs
Given the phenotypic changes we observed in cotreated GICs in vitro, we next tested whether treatment with IR and rapamycin affects in vivo differentiation potential of GICs (Supporting Information Fig. S5). Immunohistochemical analysis using Nestin and GFAP-specific antibodies revealed a significantly lower number of Nestin-positive cells in brains and a higher number of GFAP-positive cells in tumor sections in rapamycin-treated group and in cotreated group compared to the other two groups (Supporting Information Figs. S5A and 5B). To quantitatively evaluate the differentiation status of tumors in the various groups, we isolated individual intracerebral tumor cells by Fluorescence Activating Cell Sorter (FACS) sorting and determined their level of GFAP expression. Compared to the cells in untreated, IR-treated and tritreated groups, a higher percentage of cells (about threefold) expressed GFAP and did so with a higher expression intensity (Supporting Information Fig. S5C). These findings are consistent with a greater tendency for cotreated cells to undergo spontaneous differentiation in vivo compared to the cells in the other groups, thereby resulting in significantly prolonged survival in mice injected intracerebrally with cotreated cells compared to mice in the other groups.
GICs are responsible for the initiation of gliomas and show increased resistance to irradiation, a major therapeutic modality for the treatment of malignant gliomas,34 which leads GICs to be critical therapeutic targets.41, 42 Nevertheless, there have been few, if any, reports demonstrating exactly how their stemness is maintained and how they affect glioma radiosensitivity. Here, we revealed that induction of autophagy with rapamycin rendered GICs isolated from patient-derived glioma specimens more sensitive to radiation. Recently, some studies suggest that there may exist more than one marker that can prospectively enrich for a GICs population.43, 44 However, whether CD133 can serve as a universal tumor stem cells enrichment marker for all tumors has been questioned. Therefore, in our study, GICs retained their stem cell state as assessed by the markers nestin, Sox2 and Musashi, along with CD133.
Autophagy has been considered as an essential cellular process, and its importance for regulating cancer development and progression and for determining sensitivity of tumor cells to anticancer therapy has been recognized.45, 46 We and others have implicated autophagy as an alternative death pathway adopted by malignant glioma cells.12, 13, 24 In our study, we found that rapamycin, an mTOR inhibitor, sensitized GICs to the cytotoxic effect of IR. This is associated with the induction of autophagy. Apoptosis detection assay indicated that radiation sensitization in cotreated GICs was due to higher rates of apoptosis. It is important to identify the presence of biochemical switches that direct GICs toward apoptosis or autophagy. Rapamycin is the best characterized brain-accessible drug that upregulates autophagy.47 It induces autophagy by inhibiting the TOR proteins, which are well conserved from yeast to human. Our data showed that inactivation of mTOR pathway was sufficient to induce autophagy in vitro and in vivo following the administration of rapamycin and radiation. However, radiation alone did not lead to a significant change in LC3-II level. The basal activity of autophagy in the untreated GICs was very low, which was consistent with our previous study.48 To confirm the observed link between autophagy and radiosensitization, we applied rapamycin in addition to radiation on GICs to see if an enhancement of autophagic activity would augment radiation toxicity. Indeed, the capacity to achieve radiosensitization through mTOR inhibition has been shown before.14, 17 Similarly, our results indicate that rapamycin radiosensitized GICs to a greater extent than rapamycin alone or radiation alone.
In addition to its established roles in regulating nutrient supply and cell death, autophagy also functions in cellular development and differentiation. Studies of mice with global knockouts of critical autophagy genes have revealed that developmental defects occur in the absence of autophagy, particularly in the nervous system.49 These findings have given rise to the concept that autophagy may regulate development by performing critical cellular remodeling functions required for differentiation. Rather, the identification of a subset of GBMs that has this particular differentiation block will be of great utility for further understanding glioma pathophysiology and glioma stem cell biology and ultimately for designing rationale therapeutic strategies. Our recent studies have confirmed that the autophagic activity in GICs was minimal and their minimal autophagic activity was correlated with the inhibition of differentiation of GICs.48 To determine whether autophagy functions in the process of differentiation of GICs, the levels of expression of neural stem/progenitor markers and differentiation markers were examined in GICs in various groups. We found that an in vitro and in vivo induction of autophagy in GICs had profound effects on the differentiation and radiosensitivity of these cells. With induction of autophagic function, the self-renewal of GICs was inhibited, differentiation was promoted and radiosensitivity was markedly increased in vitro. Enhanced radiosensitization was accompanied by reduced clonogenic survival, decreased cell viability, increased apoptosis level and less efficient repair ability, all of which did not adversely affect the treatment tolerability in vivo. In the mice bearing GICs intracranial xenograft tumor treated with rapamycin and IR, not only autophagy activity of GICs was increased but also GICs acquired differentiation features. In vivo, these effects of an induction of autophagy in GICs translated into a decreased capacity of neurospheres to generate gliomas and a significantly longer surviving animal with increased radiosensitivity. These findings demonstrate that autophagy is critical for the differentiation of GICs from stem/progenitor cells, and in the absence of autophagic function, GICs may be blocked into a pathway of differentiation. In addition, GICs should experience a loss of autophagic function, and the autophagy defect may be associated with their poor differentiation. Moreover, our data showed that a critical function for autophagy in the differentiation of GICs determines their radiosensitivity. As has been previously reported, GICs represent a cellular population that confers glioma radioresistance and could be the source of glioma recurrence after radiation. Poorly differentiated glioma cells have greater radioresistance than well-differentiated glioma cells.34 Here, we showed that autophagy activity might facilitate the differentiation of GICs, which resulted in an increased radiosensitivity in GICs.
The main effect of rapamycin in this context is through autophagy, although the reported effects of rapamycin in glioma may result from one of the many mTOR-dependent pathways other than autophagy. To determine whether the increased radiation sensitivity in GICs resulted from autophagy induction, we used autophagy inhibitor 3-MA, which inhibits Type III PI3K, initiator of autophagy, to inhibit autophagy induction. 3-MA is commonly used to inhibit rapamycin-induced autophagy; however, 3-MA could have some autophagy-independent effects.50 As shown in the Results section, the triple treatment group (rapamycin + IR + 3-MA) did not promote differentiation of GICs and their radiosensitivity. On the basis of these results, we speculated that autophagy but not other mTOR inhibition effects governed the synergistic effect of combined IR + rapamycin. Our results indicated that mTOR inhibition promoted differentiation of GICs by inducing autophagy, and upregulation of autophagy was required in the differentiation cascades of GICs. Differential glioma cells repair radiation-induced DNA damage less effectively than GICs (undifferential glioma cells).2, 3, 34, 41 One of the limitations of rapamycin is that its targets are either indirectly or nonexclusively involved in autophagy, therefore, its use also impinges on independent or parallel systems. Although actual mechanism of this reagent in tumorigenesis is not clear, the observation that certain autophagy-inducing drug is antitumorigenic at least allows these to be considered for therapy.
Autophagy, apoptosis and differentiation are believed to be independent cellular cascades regulated by distinct signaling machineries. It remains unclear how the cell integrates inputs from multiple signaling pathways that mediate the implementation of complex cellular decisions. Understanding the intersection of the autophagy pathway with the differentiation cascades, a key feature of GICs, is a key unanswered question.
GICs and TICs are critical targets in cancer therapeutics. An understanding of the exact mechanisms involved in the regulation of differentiation of GICs and the resistance to IR would facilitate therapeutic strategy development. Our results support a notion that the differentiation cascade coincides with upregulation of autophagy, while basal autophagic activity in GICs is required in preventing GICs from differentiating. Autophagy may play a critical role in coordinating differentiation in GICs, although the mechanism for this process has not been delineated. The possibilities include aberrations in oncogenic pathways implicated in both developmental and glioma biology (e.g., EGFR, PTEN and mTOR) or other stem cell pathways. We have shown that loss or mutation of PTEN genes in GICs is one of the mechanisms leading to the scarce autophagy in GICs.48 It is worth studying the importance of autophagy in GBM, which shows some analogies with GBM, in which the occurrence of autophagic processes has been ascertained, although their role has not been clarified. Those observations, together with the data presented here, encourage us to speculate that autophagy functions in regulating differentiation of GICs and its role in radiosensitization of primary GBMs by controlling the differentiation of GICs. Such insights not only pave the way for a more thorough understanding of GICs biology but also identify autophagy as a promising molecular target and open the potential for targeted therapeutic approaches for agents that can induce terminal differentiation of GICs.
Additional Supporting Information may be found in the online version of this article.
|IJC_25975_sm_suppfig1.tif||13473K||Supporting Figure 1 Characterization of GICs from human glioblastoma specimens. (A) Representative images of neurospheres from GICs from human glioblastoma specimens. Scale bars = 100 μm. (B) Neurospheres from a expressed cancer stem/progenitor cell markers (CD133, Nestin, Sox2 and Musashi), as assessed by immunofluorescence. Scale bars = 100 μm. (C) GICs from human glioblastoma specimens exhibit multi-lineage differentiation potential. Differentiated cells derived from a single neurosphere of GICs were examined by immunofluorescent staining with specific antibodies against differentiation markers for oligodendrocytes (O4), astrocytes (GFAP), and neuronal progenitors (Map-2). DAPI staining (blue) was used to identify nuclei. Scale bars = 20 μm. Data shown were from representative experiments and error bars represent SD (performed in triplicate).|
|IJC_25975_sm_suppfig2.tif||781K||Supporting Figure 2 Effect of rapamycin on mTOR activity in GICs. Cell lysates were subjected to Western blot analysis using an antibody against phospho-Thr389-specific p70S6K. An anti-total p70S6K antibody was used to confirm equal loading of proteins.|
|IJC_25975_sm_suppfig3.tif||6923K||Supporting Figure 3 Treatment with rapamycin and IR inhibits tumorigenicity of GICs. H&E stained section five weeks after tumor implantation revealed antitumor efficacy of rapamycin in combination with IR. Tumours in untreated control mice, rapamycin-treated mice, IR-treated mice and triple-treated mice (rapamycin + IR + 3-MA) contain pleiomorphic, neoplastic and malignant, infiltrating cells and have invaded the ventricular system (a-f, i, j). Co-treated animals generated very small noninvasive tumors (g, h). Magnification was in 4× a, c, e, g, i and 10× in b, d, f, h, j. k, Kaplan-Meier curves demonstrate that combination treatment with rapamycin and IR increased the survival of mice bearing SU-2 intracranial xenografts. p > 0.05 comparing untreated control and IR; p < 0.05 comparing untreated control and rapamycin + IR; p < 0.05 comparing rapamycin and rapamycin + IR; p < 0.05 comparing IR and rapamycin + IR; p < 0.05 comparing rapamycin + IR and rapamycin + IR+ 3MA; Similar results were observed in SU-3 (data not shown).|
|IJC_25975_sm_suppfig4.tif||5091K||Supporting Figure 4 The presence of double stranded DNA breaks was assessed by immunoflourescent analysis of phosphorylated H2AX foci. (A) Representative images of phosphorylated H2AX under various treatment are shown. (B) The number of cells with nuclear foci of phosphorylated H2AX at different time points after various treatments in GICs were quantified. Descriptive statistics were graphed (mean ± s.d., n=100 cells × 4 trials; *P < 0.001). Similar results were observed in SU-2 (data not shown).|
|IJC_25975_sm_suppfig5.tif||18179K||Supporting Figure 5 (A) Representative images of immunohistochemical staining of sections of tumors in pre-transplantation treatment experiments with antibodies against GFAP and Nestin at 35 days. Scale bars = 200 μm. (B) Quantitation of nestin-positive cells in tumors derived from SU-2 cells. C, FACS analysis of GFAP expression in SU-2 derived tumor cells. Similar results were observed in SU-3 (data not shown).|
Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.