Abstract
Purpose:
The various microenvironments that exist within the brain combined with the invasive nature of glioblastoma (GBM) creates the potential for a topographic influence on tumor cell radiosensitivity. The aim of this study was to determine whether specific brain microenvironments differentially influence tumor cell radioresponse.
Methods and Materials:
GBM stem-like cells were implanted into the right striatum of nude mice. To measure radiosensitivity, proliferation status of individual tumor cells was determined according to the incorporation of 5-chloro-2’-deoxyuridine delivered at 4, 12, and 20 days after brain irradiation. As an additional measure of radiosensitivity, the percentage of human cells in the right hemisphere and the olfactory bulb were defined using digital droplet polymerase chain reaction. Targeted gene expression profiling was accomplished using NanoString analysis.
Results:
Tumor cells were detected throughout the striatum, corpus callosum, and olfactory bulb. After an initial loss of proliferating tumor cells in the corpus callosum and striatum after irradiation, there was only a minor recovery by 20 days. In contrast, the proliferation of tumor cells located in the olfactory bulb began to recover at 4 days and returned to unirradiated levels by day 12 postirradiation. The percentage of human cells in the right hemisphere and the olfactory bulb after irradiation also suggested that the tumor cells in the olfactory bulb were relatively radioresistant. Gene expression profiling identified consistent differences between tumor cells residing in the olfactory bulb and those in the right hemisphere.
Conclusions:
These results suggest that the olfactory bulb provides a radioresistant niche for GBM cells. Published by Elsevier Inc.
Summary
Glioblastoma cells residing in the olfactory bulb are radioresistant as compared to those residing in the right hemisphere.
Introduction
Glioblastoma (GBM) has long been classified as a radioresistant tumor.1–3 Whether treatment outcome reflects a homogeneous radioresponse among the cells comprising a given GBM or is dictated by a radioresistant sub-population(s) remains unclear. The presence of tumor cell subpopulations of varying radiosensitivities would have implications not only in the development of more effective radiation therapy, but also in the application of molecularly targeted radiosensitizers.
The radioresponse of brain tumor models is typically evaluated according to animal survival and more recently by using imaging techniques to define tumor growth rate, both of which describe the radiosensitivity of the tumor cell population as a whole. To identify variations in radiosensitivity among the subpopulations of a given tumor, it is necessary to evaluate radioresponse at the individual cell level. Toward this end, one approach is to deliver halogenated thymidine analogs after irradiation, followed by histochemical analysis. Halogenated thymidine analogs are incorporated into DNA during the S-phase of the cell cycle; this procedure has long been used to identify proliferating cells in normal tissue and tumors.4 With respect to radioresponse, incorporation of the halogenated thymidine analog after irradiation would be indicative of proliferation, with such cells classified as radioresistant. The histology-based identification of cells that do and do not proliferate after irradiation thus provides a strategy for defining intratumor heterogeneity in radiosensitivity.
GBMs in situ are highly migratory and extensively invasive,5 resulting in tumor cells being distributed across the multiple microenvironmental niches that exist within the brain. Consequently, a potential source of intratumor heterogeneity in radiosensitivity among the cells of a given GBM is their location at the time of irradiation. To investigate the role of specific brain microenvironments in determining tumor cell radioresponse, we used orthotopic xenografts initiated from GBM stem-like cells (GSCs). GSCs are thought to be critical to the development, maintenance, and treatment response of GBM.6 Moreover, GSC-initiated orthotopic xenografts not only replicate the genotype and phenotype of GBMs, but also their in vivo growth pattern.7,8 In the study described here, GSCs implanted into the right striatum are shown to migrate into the frontal cortex and olfactory bulb (OB). The data presented indicate that tumor cells in the OB were radioresistant compared with those in the corpus callosum (CC) and striatum. Thus, these results suggest that the murine OB provides a radioresistant niche for tumor cells.
Methods and Materials
Cell lines
The GSC lines NSC11 and NSC20 (provided by Dr Frederick Lang, MD Anderson Cancer Center) were maintained as neurospheres; CD133+ GSCs were isolated by FACS9 and met the criteria for being tumor stem-like.9,10 In vitro clonogenic survival analysis was performed as described.9
Orthotopic xenografts
GSCs (1.0 ×105), transduced to express luciferase and green fluorescent protein (GFP) with the lentivirus LVpFUGQ-UbC-ffLuc2-eGFP2 were implanted into the right striatum of 6-week-old athymic female nude mice (NCI Animal Production Program, Frederick, MD).11 Before irradiation mice were randomized according to the bioluminescent imaging signal.11 We delivered 5-chloro-2’-deoxyuridine (CldU) by daily intraperitoneal (IP) injections (100 mg/kg). Mice were euthanized 2 hours after the last injection. For irradiation, mice were anesthetized and placed in plexiglass jigs with shielding for the entire torso and critical normal structures of the head. Radiation was delivered using an X-Rad 320 X-irradiator (Precision X-Rays, Inc) at 2.9 Gy/min. All experiments were performed as approved according to the principles and procedures in the NIH Guide for Care and Use of Animals and were conducted in accordance with the Institutional Animal Care and Use Committee.
Immunohistochemical analysis
Formalin-fixed brains were embedded in paraffin and cut into 6 to 10 μm sections. The antibodies used were human SOX2 (Cell Signaling), CldU (clone BU1/75, ICR1), phospho-H2AX (Millipore), anti-rabbit-AlexaFluor488, anti-rat-AlexaFluor647, and anti-mouse-AlexaFluor555. Image analysis is described in Methods E1 (available online at https://doi.org/10.1016/j.ijrobp.2020.01.007).
Digital droplet polymerase chain reaction
Copy numbers of a human gene (myocardin-like protein 2 [MKL2]) and a mouse gene (transferrin receptor [Tfrc]) were determined by digital droplet polymerase chain reaction (ddPCR) and used to calculate the ratio of human/mouse cells (Methods E1; available online at https://doi.org/10.1016/j.ijrobp.2020.01.007).
NanoString nCounter gene expression
Using paraffin blocks generated from brain tumors on day 35 postimplant, RNA was isolated from core punches (1.5– mm diameter) with the RNease FFPE kit (Qiagen). RNA (200 ng) was analyzed using the NanoString nCounter Human PanCancer and Human Neuropath gene expression panels (NanoString Technologies) and nSolver analysis software according to NanoString guidelines.12 Differentially expressed genes (OB vs right hemisphere) were defined at a significance of P < .05 (Student’s t test). Classification into gene ontologies was performed using Panther (http://pantherdb.org/), and pathway enrichment analyses were performed using Ingenuity Pathway Analysis (IPA, Qiagen).
Results
To evaluate the radiosensitivity of human tumor cells as a function of location within the murine brain, orthotopic xenografts were generated from CD133+ NSC11 GSCs. At day 28 postimplantation, CldU was delivered for 2 to 6 days. Mice were euthanized 2 hours after the last injection, and brains were collected for immune-histochemical analysis of CldU incorporation to identify cycling cells with costaining for SOX2 to identify the NSC11 cells in the mouse background. The survival of unirradiated mice after NSC11 implantation is approximately 55 days.13 SOX2 expression was detected in the right striatum (implantation site), CC and cortex, and the left hemisphere (Fig. 1A). The specificity of SOX2 staining for human tumor cells is illustrated in Figure E1 (available online at https://doi.org/10.1016/j.ijrobp.2020.01.007). CldU+ tumor cells were detectable after 2 days of injections, with an increase in the percentage of tumor cells (SOX+) expressing CldU at 4 days. Higher magnification of SOX2+ and CldU+ cells (Fig. 1B) was used for the quantification of CldU+ tumor cells. CldU was detected in approximately 50% of tumor cells in the right striatum and CC after 2 days and increased to >80% after 6 days (Fig. 1C). These results indicate the xenografts contain a large fraction of cycling cells.
Fig. 1.
Chloro-2’-deoxyuridine (CldU) incorporation into NSC11 xenografts. On day 28 postimplantation, daily CldU treatment was initiated (100 mg/kg, intraperitoneal). Mice were euthanized 2 hours after the last injection. (A) Representative coronal images collected after 2 and 4 days of CldU delivery. (B) Representative high magnification images. (C) CldU+ tumor cells in each location (right hemisphere) as a function of the days of CldU treatment. Each point corresponds to an individual mouse with at least 200 cells analyzed (mean ±standard error of the mean for 4–5 mice).
This procedure was then applied to irradiated tumors. On day 35 postimplantation, NSC11 brain tumors received 10 Gy; on days 4, 12, and 20 after irradiation, CldU delivery was initiated (3 daily IP injections with tumors collected 2 hours after the last dose). The 4-day delay between irradiation and the first CldU injection was chosen to allow for resolution of the presumed initial transient cell cycle arrest. At the specified times after irradiation, immunohistochemical analyses of CldU incorporation and SOX2 expression was performed to determine the percentage of tumor cells that were proliferating in the CC, striatum, and OB (Fig. 2A). SOX2 staining in the sagittal section of the right hemisphere from a control mouse (Fig. 2A top panel) shows that by day 35 after implantation, NSC11 cells had migrated into the frontal cortex and to the right OB.
Fig. 2.
NSC11 cell proliferation after brain irradiation. Thirty-five days postimplant, mice received 10 Gy; chloro-2’-deoxyuridine (CldU) delivery (3 daily doses with the tumor collected 2 hours after the last dose) was initiated on the specified day after irradiation. CldU delivery was initiated for control mice on day 35. (A) Representative matched images of sagittal sections of the right hemisphere and coronal sections of the right olfactory bulb (OB) of a control mouse (top) and a mouse in which CldU delivery was initiated on day 12 after 10 Gy (bottom). (B) CldU+ tumor cells identified in the CC, STR, and OB of matched mice as a function of time after 10 Gy. Each point corresponds to an individual mouse (mean ± standard error of the mean for 3–4 mice), *P < .05 versus control. (C) Percentage of human cells in the right OB and corresponding right hemisphere as a function of time after 10 Gy. Mice were collected on day 0 (day 35 postimplant) or received 10 Gy and were collected for analysis 7 and 14 days later along with unirradiated controls. Values represent the mean ± standard error of the mean of 4 mice, *P < .05 versus day 35, Student’s t test. Abbreviations: CC = corpus callosum; GL = glomerular layer; GCL = granule cell layer; HIP = hippocampus; STR = striatum.
When CldU injections were initiated 12 days after irradiation, there was a reduction in CldU+ tumor cells in the CC and striatum (bottom panel, Fig. 2A). However, in the OB, the level of CldU+ tumor cells in the irradiated mice was similar to that in the unirradiated animals. The percentage of CldU+ tumors in each location as a function of time after 10 Gy is shown in Figure 2B. When CldU was delivered at 4 days after 10 Gy, less than 4% of tumor cells in the CC and striatum were CldU+; the levels increased slightly when CldU delivery was initiated 20 days after irradiation. In the OB, although there was considerable variability among mice at 4 days after 10 Gy, the percentage of CldU+ tumor cells was similar to controls at 12 days and increased above controls at 20 days after 10 Gy. Thus, although there was some recovery of tumor cell proliferation in the CC and striatum 20 days after irradiation, recovery of NSC11 cells in the OB was faster and more complete. These initial results suggest that tumor cells in the OB are relatively radioresistant.
As an orthogonal approach to comparing the radiosensitivity of tumor cells in the OB versus the right hemisphere, ddPCR was used to determine the number of copies of a human-specific gene (MKL2) and a mouse-specific gene (Tfrc) in each region; this provides a measure of the percentage of human cells. On days 35, 42, and 49 after implantation of NSC11 cells into the right striatum, the right OB and right hemisphere were collected by gross dissection of GFP-expressing tumor tissue; DNA was isolated from each region, and the percentage of human cells was determined. In addition, 8 mice were irradiated (10 Gy) on day 35, and their right OB and right hemisphere were collected on days 42 and 49 postimplant, respectively. The percentage of human cells increased in the unirradiated OB and right hemisphere over the 14-day analysis period (Fig. 2C), consistent with tumor cell proliferation. After 10 Gy, the percentage of human cells in the right hemisphere declined over the 14-day period. In contrast, in the OB the percentage of human cells had slightly increased at 14 days after 10 Gy. These results, consistent with those generated from the analysis of CldU+ cells, indicate that NSC11 cells in the OB are radioresistant compared with those in the right hemisphere.
As an estimation of in vivo dosimetry, γH2AX foci14 were defined in the CC, striatum, and OB. Mice were irradiated, and γH2AX foci were scored in SOX2-expressing cells in each location at 0.5 and 6 hours postirradiation. Representative images generated from the OB are shown in Figure 3A. γH2AX foci/cell levels in each location 0.5 hours after irradiation (Fig. 3B) were similar, indicating that there was no difference in the initial level of radiation-induced DSBs in OB compared with the CC or striatum and that each location received an equivalent radiation dose.
Fig. 3.
γH2AX foci in NSC11 xenografts and in vitro radiosensitivity of NSC11 glioblastoma stem-like cells isolated from olfactory bulb and right hemisphere. On day 35 postimplant of NSC11 cells, mice received 6 Gy and were collected for analysis at the indicated time points. (A) Representative images of γH2AX foci in tumor cells (SOX2+) (40 × magnification). The last column corresponds to the overlay used for automated foci counting. (B) γH2AX foci in the corpus callosum, striatum, and olfactory bulb (OB) as a function of time after 6 Gy. Each column corresponds to an individual mouse, with at least 50 cells analyzed per mouse at each location. Bars represent mean ± standard error of the mean. (C). Two NSC11 tumor-bearing mice were euthanized on day 35 postimplant. Green florescent protein fluorescent tissue was collected from the OB and right hemisphere from each mouse, disaggregated, and seeded into standard tissue culture plates in stem cell media. The resulting neurospheres grown from each location were subjected to in vitro clonogenic survival analysis. Values represent the mean ± standard deviation of 3 independent experiments.
To further investigate the microenvironment as a determinant of tumor cell radiosensitivity, neurosphere cultures were generated from the OB and the right hemisphere of 2 mice bearing NSC11 tumors (35 days postimplant) and subjected to in vitro clonogenic survival analyses.9 For each mouse, the radiation survival curves generated from NSC11 cells isolated from the OB and right hemisphere were similar (Fig. 3C). These data suggest that the radioresistance of NSC11 cells in the OB is dependent on continued input from the microenvironment.
The analysis of tumor cell proliferation based on CldU incorporation was then extended to the orthotopic xenografts initiated from the NSC20 GSCs using the same treatment protocol as for NSC11 tumors. On day 35 postimplantation, NSC20 brain tumor xenografts were exposed to 10 Gy; on days 4 and 12 after irradiation, CldU delivery was initiated (3 daily IP injections with tumors collected 2 hours after the last dose). As shown by the SOX2 expression in the control tumors (Fig. 4A), by day 35 after implantation, although some NSC20 cells remain in the striatum, many have invaded past the CC and into the cortex. The percentage of CldU+ tumor cells was similar between frontal cortex and striatum. Although the growth pattern was different from that of NSC11 in the right hemisphere, NSC20 cells also infiltrated the OB (Fig. 4A, right panel).
Fig. 4.
NSC20 cell proliferation after brain irradiation. Thirty-five days postimplant, mice received 10 Gy; chloro-2’-deoxyuridine (CldU) delivery (3 daily doses with the tumor collected 2 hours after the last dose) was initiated on the specified days after irradiation. CldU delivery was initiated for control mice on day 35. (A) Representative matched images of sagittal sections of the right hemisphere and coronal sections of the right olfactory bulb (OB) of a control mouse and a mouse in which CldU delivery was initiated on day 12 after 10 Gy. Insets: zoomed-in images illustrating cells in the right hemisphere. (B) CldU+ tumor cells identified in the RH and OB of matched mice as a function of time after 10 Gy. Each point corresponds to an individual mouse, with at least 200 cells analyzed in each location (mean ± standard error of the mean for 3–4 mice) *P < .05 versus control. (C) Percentage of human cells in the right OB and corresponding RH as a function of time after 10 Gy. Mice were collected on day 35 (day 0) or received 10 Gy and were collected for analysis 7 and 14 days later along with unirradiated controls. Values represent the mean ± standard error of the mean of 4 to 8 mice, *P < .05 versus day 35.
When CldU injections were initiated 12 days after irradiation, there was a dramatic loss of CldU+ tumor cells throughout the right hemisphere (bottom panel, Fig. 4A). However, in the OB, the level of CldU+ tumor cells in the irradiated mice was similar to that in the control animals. Quantitation of the percentage of CldU+ tumor cells after irradiation (Fig. 4B) shows a significant reduction in the right hemisphere at 4 and 12 days after 10 Gy. In the OB, although there was considerable variability among mice at 4 days, the percentage of CldU+ tumor cells returned to control levels by 12 days after 10 Gy. These data indicate that NSC20 cells in the OB are relatively radioresistant compared with those in the right hemisphere. As for NSC11 tumors, ddPCR was used to determine the percentage of human cells in the OB and right hemisphere after irradiation. After 10 Gy, the percentage of human cells in the right hemisphere declined over the 14-day analysis period, whereas the percentage of human cells in the OB slightly increased. These results, consistent with the CldU analysis, indicate that NSC20 cells in the OB are relatively radioresistant compared with those in the right hemisphere.
To compare the molecular phenotype of the NSC11 tumor cells residing in the OB and the right hemisphere, targeted gene expression profiling was performed on RNA isolated from FFPE samples using the NanoString nCounter Analysis System, which provided a quantitative expression analysis of 1385 genes. Comparison of gene expression levels in the OB versus right hemisphere identified 233 differentially expressed genes (P < .05) (Fig. E2A, Table E1; available online at https://doi.org/10.1016/j.ijrobp.2020.01.007). These significant genes were organized into gene ontologies, and the classification of “GO biological process complete” was used to annotate the pathways. Figure 5 lists examples of the significantly overexpressed pathways that were nested within Cellular Process to Chemical Synaptic Transmission (FDR range, 2.95E-15 to 3.36E-25). This gene list was interrogated using IPA. The majority of genes (158 of 233) were present under the Kyoto Encyclopedia of Genes and Genomes heading Nervous System Development and Function with 49 subheadings related to the nervous system, each of which was significant (P < 10E-13). The networked genes within these headings (n = 135) are shown in Figure 6A, with most genes increased (red) in the tumors harvested from the OB compared with the right hemisphere. Within this network are multiple genes that correspond to calcium metabolism, glutamate receptors, and BMPs.
Fig. 5.
Biological/biochemical functions of the genes differentially expressed in NSC11 and NSC20 cells growing in the olfactory bulb versus the right hemisphere.
Fig. 6.
Gene expression for tumor cells growing in the olfactory bulb versus the right hemisphere. Ingenuity Pathway Analysis networked genes under the Kyoto Encyclopedia of Genes and Genomes heading Nervous System Development and Function for (A) NSC11 and (B) NSC20. Shades of red refer to upregulated genes in the olfactory bulb and green to downregulated genes. The color intensity of a molecule indicates the degree of increased or decreased enrichment. Other colors indicate the presence (gray) or absence (white) of a given gene.
Performing the same gene expression analysis on NSC20 cells in the OB versus the right hemisphere identified 199 significant genes (Fig. E2B, Table E2; available online at https://doi.org/10.1016/j.ijrobp.2020.01.007). The differentially expressed genes were distributed to a set of GO pathways associated with neuronal function that were similar to that of the NSC11 tumors, with all but Synaptic Transmission, glutamatergic being significant (Fig. 5). Genes from the NSC20 tumors in the OB also sorted to signal transduction (GO:0007165) more significantly than did NSC11. When the NSC20 outliers were subjected to IPA, the majority of genes also appeared under the Kyoto Encyclopedia of Genes and Genomes heading Nervous System Development and Function. The networked genes (n = 63) within this heading are shown in Figure 6B. The genes that were differentially expressed (n = 34) between tumor cells in the OB and right hemisphere for both NSC20 and NSC11 included BMPs, calcium metabolism, and glutamate receptors (Fig. E3; available online at https://doi.org/10.1016/j.ijrobp.2020.01.007).
Discussion
The assortment of microenvironments that exist within the brain, combined with the invasive nature of GBM, creates the potential for a topographic influence on tumor cell biology. In experimental models, GSCs have been reported to be preferentially located in a number of niches, including perivascular and perineuronal.15 Although noteworthy with respect to the fundamental biology of GBM, whether a given location preferentially affects tumor cell radiosensitivity has yet to be determined. To address this issue, we used GSCs that after orthotopic implantation migrate throughout the mouse brain. Results generated from 2 experimental approaches showed that tumor cells in the OB are radioresistant compared with those in the right hemisphere. These data indicate that the murine OB provides a radioresistant niche for tumor cells. More broadly, these results illustrate the potential for intratumor heterogeneity in GBM radiosensitivity.
Whether the OB in the human brain provides an analogous niche is unclear. However, defining the molecular and cellular processes that mediate radioresistance in the murine OB may generate a better understanding of the mechanisms responsible for the radioresistance of GBM in general. Along these lines, recent studies have identified a direct interaction between neurons and glioma cells, which, based on single cell transcriptome analysis, resulted in the modification of glioma gene expression and, moreover, were implicated as a source of tumor progression.16,17 The murine OB is highly enriched in neurons and interneurons that provide an environment that may promote neuron–glioma cell interactions. Accordingly, as recently reported for single cell transcriptomes,16,17 targeted expression analysis comparing the NSC11 and NSC20 cells in the murine OB and right hemisphere showed that the tumor cells residing in the OB preferentially expressed genes related to neuronal function. Thus, although defining the molecular and cellular processes that mediate radioresponse of tumor cells in murine OB may provide an opportunity to investigate a radioresistant subpopulation, it may also lead to a better understanding of the mechanisms responsible for the radioresistance of GBM in general.
Supplementary Material
Acknowledgments—
All microscopy was performed with the aid of Ross Lake of the Laboratory of Genitourinary Cancer Pathogenesis Microscopy Core Facility, NCI, NIH.
This study was funded by the Division of Basic Sciences, Intramural Program, National Cancer Institute (Z1ABC011372). This project has also been funded in part with federal funds from the National Cancer Institute, National Institutes of Health, under contract no. HHSN261200800001E. The content of this publication does not necessarily reflect the views or policies of the Department of Health and Human Services, nor does mention of trade names, commercial products, or organizations imply endorsement by the US Government.
Footnotes
Disclosures: none.
Research data are available and will be shared upon request to the corresponding author.
Supplementary material for this article can be found at https://doi.org/10.1016/j.ijrobp.2020.01.007.
References
- 1.Stupp R, Mason WP, van den Bent MJ, et al. Radiotherapy plus concomitant and adjuvant temozolomide for glioblastoma. N Engl J Med 2005;352:987–996. [DOI] [PubMed] [Google Scholar]
- 2.Chan JL, Lee SW, Fraass BA, et al. Survival and failure patterns of high-grade gliomas after three-dimensional conformal radiotherapy. J Clin Oncol 2002;20:1635–1642. [DOI] [PubMed] [Google Scholar]
- 3.Laperriere N, Zuraw L, Cairncross G. Radiotherapy for newly diagnosed malignant glioma in adults: A systematic review. Radiother Oncol 2002;64:259–273. [DOI] [PubMed] [Google Scholar]
- 4.Vega CJ, Peterson DA. Stem cell proliferative history in tissue revealed by temporal halogenated thymidine analog discrimination. Nat Methods 2005;2:167–169. [DOI] [PubMed] [Google Scholar]
- 5.Giese A, Bjerkvig R, Berens ME, et al. Cost of migration: Invasion of malignant gliomas and implications for treatment. J Clin Oncol 2003; 21:1624–1636. [DOI] [PubMed] [Google Scholar]
- 6.Venere M, Fine HA, Dirks PB, et al. Cancer stem cells in gliomas: Identifying and understanding the apex cell in cancer’s hierarchy. Glia 2011;59:1148–1154. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Lee J, Kotliarova S, Kotliarov Y, et al. Tumor stem cells derived from glioblastomas cultured in bFGF and EGF more closely mirror the phenotype and genotype of primary tumors than do serum-cultured cell lines. Cancer Cell 2006;9:391–403. [DOI] [PubMed] [Google Scholar]
- 8.Vescovi AL, Galli R, Reynolds BA. Brain tumour stem cells. Nat Rev Cancer 2006;6:425–436. [DOI] [PubMed] [Google Scholar]
- 9.McCord AM, Jamal M, Shankavaram UT, Lang FF, Camphausen K, Tofilon PJ. Physiologic oxygen concentration enhances the stem-like properties of CD133+ human glioblastoma cells in vitro. Mol Cancer Res 2009;7:489–497. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Jamal M, Rath BH, Williams ES, Camphausen K, Tofilon PJ. Microenvironmental regulation of glioblastoma radioresponse. Clin Cancer Res 2010;16:6049–6059. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Jamal M, Rath BH, Tsang PS, Camphausen K, Tofilon PJ. The brain microenvironment preferentially enhances the radioresistance of CD133(+) glioblastoma stem-like cells. Neoplasia 2012;14:150–158. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Kulkarni MM. Digital multiplexed gene expression analysis using the NanoString nCounter system. Curr Protoc Mol Biol 2011. Chapter 25: Unit25B.10. [DOI] [PubMed] [Google Scholar]
- 13.Timme CR, Rath BH, O’Neill JW, Camphausen K, Tofilon PJ. DNA-PK inhibitor VX-984 enhances the radiosensitivity of glioblastoma cells grown in vitro and as orthotopic xenografts. Mol Cancer Ther 2018;17:1207–1216. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Bonner WM, Redon CE, Dickey JS, et al. GammaH2AX and cancer. Nat Rev Cancer 2008;8:957–967. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Silver DJ, Lathia JD. Revealing the glioma cancer stem cell interactome, one niche at a time. J Pathol 2018;244:260–264. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Venkataramani V, Tanev DI, Strahle C, et al. Glutamatergic synaptic input to glioma cells drives brain tumour progression. Nature2019;573:532–538. [DOI] [PubMed] [Google Scholar]
- 17.Venkatesh HS, Morishita W, Geraghty AC, et al. Electrical and synaptic integration of glioma into neural circuits. Nature 2019;573:539–545. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.