The role of CXCR3/LRP1 cross-talk in the invasion of primary brain tumors

CXCR3 plays important roles in angiogenesis, inflammation, and cancer. However, the precise mechanism of regulation and activity in tumors is not well known. We focused on CXCR3-A conformation and on the mechanisms controlling its activity and trafficking and investigated the role of CXCR3/LRP1 cross talk in tumor cell invasion. Here we report that agonist stimulation induces an anisotropic response with conformational changes of CXCR3-A along its longitudinal axis. CXCR3-A is internalized via clathrin-coated vesicles and recycled by retrograde trafficking. We demonstrate that CXCR3-A interacts with LRP1. Silencing of LRP1 leads to an increase in the magnitude of ligand-induced conformational change with CXCR3-A focalized at the cell membrane, leading to a sustained receptor activity and an increase in tumor cell migration. This was validated in patient-derived glioma cells and patient samples. Our study defines LRP1 as a regulator of CXCR3, which may have important consequences for tumor biology.

The angiogenic switch leads to a metabolic shift in human glioblastoma

Background

Invasion and angiogenesis are major hallmarks of glioblastoma (GBM) growth. While invasive tumor cells grow adjacent to blood vessels in normal brain tissue, tumor cells within neovascularized regions exhibit hypoxic stress and promote angiogenesis. The distinct microenvironments likely differentially affect metabolic processes within the tumor cells.

Methods

In the present study, we analyzed gene expression and metabolic changes in a human GBM xenograft model that displayed invasive and angiogenic phenotypes. In addition, we used glioma patient biopsies to confirm the results from the xenograft model.

Results

We demonstrate that the angiogenic switch in our xenograft model is linked to a proneural-to-mesenchymal transition that is associated with upregulation of the transcription factors BHLHE40, CEBPB, and STAT3. Metabolic analyses revealed that angiogenic xenografts employed higher rates of glycolysis compared with invasive xenografts. Likewise, patient biopsies exhibited higher expression of the glycolytic enzyme lactate dehydrogenase A and glucose transporter 1 in hypoxic areas compared with the invasive edge and lower-grade tumors. Analysis of the mitochondrial respiratory chain showed reduction of complex I in angiogenic xenografts and hypoxic regions of GBM samples compared with invasive xenografts, nonhypoxic GBM regions, and lower-grade tumors. In vitro hypoxia experiments additionally revealed metabolic adaptation of invasive tumor cells, which increased lactate production under long-term hypoxia.

Conclusions

The use of glycolysis versus mitochondrial respiration for energy production within human GBM tumors is highly dependent on the specific microenvironment. The metabolic adaptability of GBM cells highlights the difficulty of targeting one specific metabolic pathway for effective therapeutic intervention.

Trifluoperazine, a novel autophagy inhibitor, increases radiosensitivity in glioblastoma by impairing homologous recombination

BACKGROUND

Resistance to adjuvant radiotherapy is a major cause of treatment failure in patients with glioblastoma (GBM). Autophagy inhibitors have been shown to enhance the efficacy of radiotherapy for certain solid tumors. However, current inhibitors do not penetrate the blood-brain-barrier (BBB). Here, we assessed the radiosensitivity effects of the antipsychotic drug trifluoperazine (TFP) on GBM in vitro and in vivo.

METHODS

U251 and U87 GBM cell lines as well as GBM cells from a primary human biopsy (P3), were used in vitro and in vivo to evaluate the efficacy of TFP treatment. Viability and cytotoxicity was evaluated by CCK-8 and clonogenic formation assays. Molecular studies using immunohistochemistry, western blots, immunofluorescence and qPCR were used to gain mechanistic insight into the biological activity of TFP. Preclinical therapeutic efficacy was evaluated in orthotopic xenograft mouse models.

RESULTS

IC50 values of U251, U87 and P3 cells treated with TFP were 16, 15 and 15.5 μM, respectively. TFP increased the expression of LC3B-II and p62, indicating a potential disruption of autophagy flux. These results were further substantiated by a decreased Lysotracker Red uptake, indicating impaired acidification of the lysosomes. We show that TFP and radiation had an additive effect when combined. This effect was in part due to impaired TFP-induced homologous recombination. Mechanistically we show that down-regulation of cathepsin L might explain the radiosensitivity effect of TFP. Finally, combining TFP and radiation resulted in a significant antitumor effect in orthotopic GBM xenograft models.

CONCLUSIONS

This study provides a strong rationale for further clinical studies exploring the combination therapy of TFP and radiation to treat GBM patients.

The anti-vascular endothelial growth factor receptor-1 monoclonal antibody D16F7 inhibits invasiveness of human glioblastoma and glioblastoma stem cells

BACKGROUND

Glioblastoma (GBM) is a highly migratory, invasive, and angiogenic brain tumor. Like vascular endothelial growth factor-A (VEGF-A), placental growth factor (PlGF) promotes GBM angiogenesis. VEGF-A is a ligand for both VEGF receptor-1 (VEGFR-1) and VEGFR-2, while PlGF interacts exclusively with VEGFR-1. We recently generated the novel anti-VEGFR-1 monoclonal antibody (mAb) D16F7 that diminishes VEGFR-1 homodimerization/activation without affecting VEGF-A and PlGF binding.

METHODS

In the present study, we evaluated the expression of VEGFR-1 in human GBM tissue samples (n = 42) by immunohistochemistry, in cell lines (n = 6) and GBM stem cells (GSCs) (n = 18) by qRT-PCR and/or western blot analysis. In VEGFR-1 positive GBM or GSCs we also analyzed the ability of D16F7 to inhibit GBM invasiveness in response to VEGF-A and PlGF.

RESULTS

Most of GBM specimens stained positively for VEGFR-1 and all but one GBM cell lines expressed VEGFR-1. On the other hand, in GSCs the expression of the receptor was heterogeneous. D16F7 reduced migration and invasion of VEGFR-1 positive GBM cell lines and patient-derived GSCs in response to VEGF-A and PlGF. Interestingly, this effect was also observed in VEGFR-1 positive GSCs transfected to over-express wild-type EGFR (EGFRwt⁺) or mutant EGFR (ligand binding domain-deficient EGFRvIII⁺). Furthermore, D16F7 suppressed intracellular signal transduction in VEGFR-1 over-expressing GBM cells by reducing receptor auto-phosphorylation at tyrosine 1213 and downstream Erk1/2 activation induced by receptor ligands.

CONCLUSION

The results from this study suggest that VEGFR-1 is a relevant target for GBM therapy and that D16F7-derived humanized mAbs warrant further investigation.

Abstract 2865: The peptide transporter K16ApoE increases drug delivery across the blood brain barrier in an experimental animal model of melanoma brain metastases

Introduction: Patients with brain metastases await a dismal prognosis. Regardless of the continuous progress in drug development, a major problem is the delivery of drugs across the blood brain barrier (BBB) and into the metastatic neoplasms. The BBB excludes almost all compounds, in particular highly charged, hydrophilic or large compounds, and most of the current chemotherapeutic agents are thus unable to penetrate the BBB. Varying strategies to transiently open the BBB have been studied previously. Here, we describe a peptide transporter comprising 16 lysine residues and 20 amino acid residues corresponding to the low density lipoprotein receptor (LDLR) binding domain of apolipoprotein E (ApoE). We show that the peptide (K16ApoE) is able to transiently open the BBB for drug-delivery into experimental brain metastases.

Experimental procedures: A systemic study of the ability of the peptide to open the BBB was conducted by dynamic contrast enhanced magnetic resonance imaging (DCE MRI) in nonobese diabetic/severe combined (nod/scid) mice. The BBB permeability was studied after administering 200 μg of the peptide intravenously. Further, cellular effects after treatment with the peptide was investigated in vitro using confocal microscopy, flow cytometry and impedance experiments. The biodistribution of the peptide was studied in blood plasma and several organs using 125I labeled K16ApoE. Finally, a treatment study was initiated, treating the animals with the peptide in combination with the B-RAF inhibitor Dabrafenib, only Dabrafenib or vehicle.

Summary: After injecting the K16ApoE peptide into the mice, a transient opening of the BBB for up to 4 hours was clearly demonstrated by DCE-MRI. Microscopy showed that the peptide disrupted brain endothelial cell monolayers, reducing the barrier properties of the cells. The impedance experiments displayed that the permeability through endothelial cell barriers was increased after treatment with K16ApoE, and a dose-dependent cell death pattern was observed at higher concentrations of K16ApoE.The peptide did not affect endothelial cell tight junctions. The biodistribution study showed that the peptide was eliminated from blood plasma in less than five minutes through the kidneys. The treatment study displayed that the group of animals receiving K16ApoE followed by Dabrafenib had smaller tumor volumes than the other two animal groups.

Conclusions: We have shown that the peptide opens the BBB and facilitates a therapeutic window of 4 hours. The peptide did in combination with Dabrafenib decrease the number of experimental brain metastases in our studies. Thus, the current strategy could also have the potential to improve the treatment of patients with brain metastatic disease.

Citation Format: Synnøve Nymark Aasen, Heidi Espedal, Olivier Keunen, Christopher Florian Holte, Habib Baghirov, Rolf Bjerkvig, Tine Veronica Karlsen, Olav Tenstad, Dag Erlend Olberg, Gobinda Sarkar, Robert B Jenkins, Frits Thorsen. The peptide transporter K16ApoE increases drug delivery across the blood brain barrier in an experimental animal model of melanoma brain metastases [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2017; 2017 Apr 1-5; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2017;77(13 Suppl):Abstract nr 2865. doi:10.1158/1538-7445.AM2017-2865

Cathepsin L silencing increases As2O3 toxicity in malignantly transformed pilocytic astrocytoma MPA58 cells by activating caspases 3/7

Low-grade, pilocytic astrocytomas are treated by resection, but additional therapy is necessary for those tumors with anaplastic features. Arsenic trioxide (As₂O₃) is emerging as an effective chemotherapeutic agent for treatment of malignant glioblastoma multiforme, where Cathepsin L silencing enables lower, less harmful As₂O₃ concentrations to achieve the desired cytotoxic effect. Here, we evaluated the effects of As₂O₃ combined with stable Cathepsin L shRNA silencing on cell viability/metabolic activity, and apoptosis in primary cultures of recurrent malignantly transformed pilocytic astrocytoma (MPA). These cells expressed high Cathepsin L levels, and when grown as monolayers and spheroids, they were more resistant to As₂O₃ than the U87MG glioblastoma cell line. Caspases 3/7 activity in MPA58 spheroids was not significantly affected by As₂O₃, possibly due to higher chemoresistance of primary biopsy tissue of less malignant astrocytoma versus the malignant U87MG cell line. However, As₂O₃ treatment was cytotoxic to MPA spheroids after silencing of Cathepsin L expression. While Cathepsin L silencing only slightly decreased the live/dead cell ratio in As₂O₃-treated MPA-si spheroids under our experimental conditions, there was an increase in As₂O₃-mediated apoptosis in MPA-si spheroids, as indicated by elevated caspases 3/7 activity. Therefore, Cathepsin L silencing by gene manipulation can be applied when a more aggressive approach is needed in treatment of pilocytic astrocytomas with anaplastic features.

Surgical debulking promotes recruitment of macrophages and triggers glioblastoma phagocytosis in combination with CD47 blocking immunotherapy

Surgical resection is a standard component of treatment in the clinical management of patients with glioblastoma multiforme (GBM). However, experimental therapies are rarely investigated in the context of tumor debulking in preclinical models. Here, a surgical debulking GBM xenograft model was developed in nude rats, and was used in combination with CD47 blocking immunotherapy, a novel treatment strategy that triggers phagocytosis of tumor cells by macrophages in diverse cancer types including GBM. Orthotopic patient-derived xenograft tumors expressing CD47 were resected at 4 weeks after implantation and immediately thereafter treated with anti-CD47 or control antibodies injected into the cavity. Debulking prolonged survival (median survival, 68.5 vs 42.5 days, debulking and non-debulking survival times, respectively; n = 6 animals/group; P = 0.0005). Survival was further improved in animals that underwent combination treatment with anti-CD47 mAbs (median survival, 81.5 days vs 69 days, debulking + anti-CD47 vs debulking + control IgG, respectively; P = 0.0007). Immunohistochemistical staining of tumor sections revealed an increase in recruitment of cells positive for CD68, a marker for macrophages/immune cell types, to the surgical site (50% vs 10%, debulking vs non-debulking, respectively). Finally, analysis of tumor protein lysates on antibody microarrays demonstrated an increase in pro-inflammatory cytokines, such as CXCL10, and a decrease in angiogenic proteins in debulking + anti-CD47 vs non-debulking + IgG tumors. The results indicated that surgical resection combined with anti-CD47 blocking immunotherapy promoted an inflammatory response and prolonged survival in animals, and is therefore an attractive strategy for clinical translation.

Rapid Conversion of Mutant IDH1 from Driver to Passenger in a Model of Human Gliomagenesis

Missense mutations in the active site of isocitrate dehydrogenase 1 (IDH1) biologically and diagnostically distinguish low-grade gliomas and secondary glioblastomas from primary glioblastomas. IDH1 mutations lead to the formation of the oncometabolite 2-hydroxyglutarate (2-HG) from the reduction of α-ketoglutarate (α-KG), which in turn facilitates tumorigenesis by modifying DNA and histone methylation as well blocking differentiation processes. Although mutant IDH1 expression is thought to drive the gliomagenesis process, the extent to which it remains a viable therapeutic target remains unknown. To address this question, we exposed immortalized (p53/pRb deficient), untransformed human astrocytes to the mutant IDH1 inhibitor AGI-5198 prior to, concomitant with, or at intervals after, introduction of transforming mutant IDH1, then measured effects on 2-HG levels, histone methylation (H3K4me3, H3K9me2, H3K9me3, or H3K27me3), and growth in soft agar. Addition of AGI-5198 prior to, or concomitant with, introduction of mutant IDH1 blocked all mutant IDH1-driven changes, including cellular transformation. Addition at time intervals as short as 4 days following introduction of mutant IDH1 also suppressed 2-HG levels, but had minimal effects on histone methylation, and lost the ability to suppress clonogenicity in a time-dependent manner. Furthermore, in two different models of mutant IDH1-driven gliomagenesis, AGI-5198 exposures that abolished production of 2-HG also failed to decrease histone methylation, adherent cell growth, or anchorage-independent growth in soft agar over a prolonged period. These studies show although mutant IDH1 expression drives gliomagenesis, mutant IDH1 itself rapidly converts from driver to passenger.

IMPLICATIONS

Agents that target mutant IDH may be effective for a narrow time and may require further optimization or additional therapeutics in glioma. Mol Cancer Res; 14(10); 976-83. ©2016 AACR.

Abstract 4413: Thrombospondin-1 is a master regulator of glioblastoma vascularization and infiltration

Glioblastoma are heterogeneous tumours composed of different subpopulations of cells with different behavioral characteristics. Areas of highly angiogenic cells co-exist with populations of non-angiogenic but highly invasive and infiltrative cells. This heterogeneity is a key clinical challenge in glioma treatment.

We used a large scale RNA sequencing approach to investigate the molecular components which differentiate infiltrative from angiogenic cells, in a Patient Derived Xenograft (PDX) mouse model. Our bioinformatic analysis revealed TGFβ1 to be a master regulator of tumor development, and Thrombospondin-1 (Tsp1) to be up regulated in infiltrative areas as compared to angiogenic area. Thrombospondins are known as anti-angiogenic factors, however the full roles of these multi domain proteins in tumor development remain to be elucidated.

We found Tsp1 expression to be upregulated in grade IV-GBM and in silico in the GBM mesenchymal subclass. It is expressed not only in tumor cells, but also in tumor blood vessel endothelial cells (ECs). TGFβ1 transcriptionally regulated Tsp1 via Smad1 and Smad3. However, contrary to previous results, we found that Tsp1 was not involved in TGFβ1-activation in tumor cells. We showed that Tsp1 regulates cell migration and invasion both in vitro and in vivo. Inhibition of Tsp1 expression in vivo correlated with increased tumor vascularization in both the chick CAM assay and the PDX mouse model.

Anti-angiogenic treatments in the PDX mouse model leads to increased tumor hypoxia and invasive tumor cell behavior, as described in our previous work. In this context, both TGFβ1 and Tsp1 were upregulated in tumor cells. Downregulation of tumor-derived Tsp1 by shRNA in the presence of anti-angiogenic therapy led to reduced tumor growth and invasion in vivo. Finally, peptide-mediated inhibition of Tsp1 activity demonstrated that Tsp1/CD47 interaction is involved in the invasive capacity of GBM cells.

Taken together, our data suggest that Tsp1 inhibition may be a promising therapeutic approach to limit tumor infiltration induced by treatment with anti-angiogenic agents.

Citation Format: Thomas Daubon, Celine Leon, Kim Clarke, Mathilde Poulet, Hrvoje Miletic, Francesco Falciani, Rolf Bjerkvig, Andreas Bikfalvi. Thrombospondin-1 is a master regulator of glioblastoma vascularization and infiltration. [abstract]. In: Proceedings of the 107th Annual Meeting of the American Association for Cancer Research; 2016 Apr 16-20; New Orleans, LA. Philadelphia (PA): AACR; Cancer Res 2016;76(14 Suppl):Abstract nr 4413.

Abstract 1343: P32-targeting TT1 peptide delivers nanoparticles to intracranial glioblastomas

Targeted delivery of cancer therapeutics using affinity ligands can dramatically improve antitumor efficacy. Over the years a number of homing peptides that upon systemic injection accumulate in solid tumors have been identified by in vivo peptide phage display. In a quest to find homing peptides optimally suited for drug delivery to high-grade gliomas, our laboratories are using advanced mouse models of glioblastoma (GBM) to systematically audit known tumor-homing peptides and to perform new in vivo screens using peptide phage libraries.

P32 is a mitochondrial chaperone that is aberrantly expressed on the cell surface in activated malignant and stromal cells in tumors. P32 is a receptor for widely used LypP-1 peptide and for recently identified TT1 peptide. Here we show that iron oxide nanoworms (IONW) functionalized with linear TT1 peptide (CKRGARST) strongly home to intracranial GBMs grafted in immune deficient mice. IONW are paramagnetic nanoparticles that are PEGylated to extend blood half-life, and have, because of their elongated shape, more effective targeting properties than spherical nanoparticles. Five hours after intravenous injection of IONW (7.5 mg/kg), macroscopic fluorescence imaging demonstrated robust homing of TT1-IONW in GBMs of murine origin (WT GBM and VEGF KO GBM from the G. Berger lab) and in a patient-derived glioma model (P13 model from the Bjerkvig lab). Confocal microscopy confirmed the presence of TT1 but not control IONW in gliomas, with TT1-IONW signal showing partial overlap with blood- and lymphatic vessel markers (CD31 and LYVE-1) in WT GBM and P13 gliomas, whereas lower level of colocalization with these markers was detected for mouse GBM not expressing VEGF. In addition, moderate colocalization between TT1-IONW and the macrophage marker CD11b was detected in the P13 tumors. Detailed phenotyping and functional characterization of TT1-positive macrophages is ongoing.

Our data suggest that TT1 peptide has potential applications as a glioma-targeting vehicle. Currently, we are evaluating TT1-targeted IONWs as a contrast agent for glioma MRI and as carriers for cytotoxic compounds.

Citation Format: Pille Säälik, Hedi Hunt, Allan Tobi, Anne-Mari Anton Willmore, Kadri Toome, Shweta Sharma, Ramana Kotamraju, Gabriele Bergers, Rolf Bjerkvig, Erkki Ruoslahti, Tambet Teesalu, Tambet Teesalu. P32-targeting TT1 peptide delivers nanoparticles to intracranial glioblastomas. [abstract]. In: Proceedings of the 107th Annual Meeting of the American Association for Cancer Research; 2016 Apr 16-20; New Orleans, LA. Philadelphia (PA): AACR; Cancer Res 2016;76(14 Suppl):Abstract nr 1343.

EGFRvIII mutations can emerge as late and heterogenous events in glioblastoma development and promote angiogenesis through Src activation

BACKGROUND

Amplification of the epidermal growth factor receptor (EGFR) and its mutant EGFRvIII are among the most common genetic alterations in glioblastoma (GBM), the most frequent and most aggressive primary brain tumor.

METHODS

In the present work, we analyzed the clonal evolution of these major EGFR aberrations in a small cohort of GBM patients using a unique surgical multisampling technique. Furthermore, we overexpressed both receptors separately and together in 2 patient-derived GBM stem cell lines (GSCs) to analyze their functions in vivo in orthotopic xenograft models.

RESULTS

In human GBM biopsies, we identified EGFR amplification as an early event because EGFRvIII mutations emerge from intratumoral heterogeneity later in tumor development. To investigate the biological relevance of this distinct developmental pattern, we established experimental model systems. In these models, EGFR⁺ tumor cells showed activation of classical downstream signaling pathways upon EGF stimulation and displayed enhanced invasive growth without evidence of angiogenesis in vivo. In contrast, EGFRvIII⁺ tumors were driven by activation of the prototypical Src family kinase c-Src that promoted VEGF secretion leading to angiogenic tumor growth.

CONCLUSIONS

The presented work shows that sequential EGFR amplification and EGFRvIII mutations might represent concerted evolutionary events that drive the aggressive nature of GBM by promoting invasion and angiogenesis via distinct signaling pathways. In particular, c-SRC may be an attractive therapeutic target for tumors harboring EGFRvIII as we identified this protein specifically mediating angiogenic tumor growth downstream of EGFRvIII.

Transmembrane protein CD9 is glioblastoma biomarker, relevant for maintenance of glioblastoma stem cells

The cancer stem cell model suggests that glioblastomas contain a subpopulation of stem-like tumor cells that reproduce themselves to sustain tumor growth. Targeting these cells thus represents a novel treatment strategy and therefore more specific markers that characterize glioblastoma stem cells need to be identified. In the present study, we performed transcriptomic analysis of glioblastoma tissues compared to normal brain tissues revealing sensible up-regulation of CD9 gene. CD9 encodes the transmembrane protein tetraspanin which is involved in tumor cell invasion, apoptosis and resistance to chemotherapy. Using the public REMBRANDT database for brain tumors, we confirmed the prognostic value of CD9, whereby a more than two fold up-regulation correlates with shorter patient survival. We validated CD9 gene and protein expression showing selective up-regulation in glioblastoma stem cells isolated from primary biopsies and in primary organotypic glioblastoma spheroids as well as in U87-MG and U373 glioblastoma cell lines. In contrast, no or low CD9 gene expression was observed in normal human astrocytes, normal brain tissue and neural stem cells. CD9 silencing in three CD133+ glioblastoma cell lines (NCH644, NCH421k and NCH660h) led to decreased cell proliferation, survival, invasion, and self-renewal ability, and altered expression of the stem-cell markers CD133, nestin and SOX2. Moreover, CD9-silenced glioblastoma stem cells showed altered activation patterns of the Akt, MapK and Stat3 signaling transducers. Orthotopic xenotransplantation of CD9-silenced glioblastoma stem cells into nude rats promoted prolonged survival. Therefore, CD9 should be further evaluated as a target for glioblastoma treatment.

Experimental therapies: gene therapies and oncolytic viruses

Glioblastoma is the most common and aggressive primary brain tumor in adults. Over the past three decades, the overall survival time has only improved by a few months, therefore novel alternative treatment modalities are needed to improve clinical management strategies. Such strategies should ultimately extend patient survival. At present, the extensive insight into the molecular biology of gliomas, as well as into genetic engineering techniques, has led to better decision processes when it comes to modifying the genome to accommodate suicide genes, cytokine genes, and tumor suppressor genes that may kill cancer cells, and boost the host defensive immune system against neoantigenic cytoplasmic and nuclear targets. Both nonreplicative viral vectors and replicating oncolytic viruses have been developed for brain cancer treatment. Stem cells, microRNAs, nanoparticles, and viruses have also been designed. These have been armed with transgenes or peptides, and have been used both in laboratory-based experiments as well as in clinical trials, with the aim of improving selective killing of malignant glioma cells while sparing normal brain tissue. This chapter reviews the current status of gene therapies for malignant gliomas and highlights the most promising viral and cell-based strategies under development.

Glutamine synthetase activity fuels nucleotide biosynthesis and supports growth of glutamine-restricted glioblastoma

L-Glutamine (Gln) functions physiologically to balance the carbon and nitrogen requirements of tissues. It has been proposed that in cancer cells undergoing aerobic glycolysis, accelerated anabolism is sustained by Gln-derived carbons, which replenish the tricarboxylic acid (TCA) cycle (anaplerosis). However, it is shown here that in glioblastoma (GBM) cells, almost half of the Gln-derived glutamate (Glu) is secreted and does not enter the TCA cycle, and that inhibiting glutaminolysis does not affect cell proliferation. Moreover, Gln-starved cells are not rescued by TCA cycle replenishment. Instead, the conversion of Glu to Gln by glutamine synthetase (GS; cataplerosis) confers Gln prototrophy, and fuels de novo purine biosynthesis. In both orthotopic GBM models and in patients, (13)C-glucose tracing showed that GS produces Gln from TCA-cycle-derived carbons. Finally, the Gln required for the growth of GBM tumours is contributed only marginally by the circulation, and is mainly either autonomously synthesized by GS-positive glioma cells, or supplied by astrocytes.