syddansk universitet metabolic challenge of proneural ... · metabolic challenge of glioblastoma...

Syddansk Universitet

Metabolic challenge of proneural glioblastoma cancer stem cell lines does not induce amesenchymal transformation in vitroBeier, Dagmar; Rasmussen, Tine; Munthe, Sune; Kristensen, Bjarne Winther; Beier,Christoph PatrickPublished in:International Journal of Clinical and Experimental Pathology

Publication date:2016

Document versionFinal published version

Document licenseCC BY-NC

Citation for pulished version (APA):Beier, D., Rasmussen, T., Munthe, S., Kristensen, B. W., & Beier, C. P. (2016). Metabolic challenge of proneuralglioblastoma cancer stem cell lines does not induce a mesenchymal transformation in vitro. International Journalof Clinical and Experimental Pathology, 9(7), 6724-6737.

General rightsCopyright and moral rights for the publications made accessible in the public portal are retained by the authors and/or other copyright ownersand it is a condition of accessing publications that users recognise and abide by the legal requirements associated with these rights.

• Users may download and print one copy of any publication from the public portal for the purpose of private study or research. • You may not further distribute the material or use it for any profit-making activity or commercial gain • You may freely distribute the URL identifying the publication in the public portal ?

Take down policyIf you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediatelyand investigate your claim.

Download date: 19. Apr. 2017

Int J Clin Exp Pathol 2016;9(7):6724-6737www.ijcep.com /ISSN:1936-2625/IJCEP0027907

Original ArticleMetabolic challenge of proneural glioblastoma cancer stem cell lines does not induce a mesenchymal transformation in vitro

Dagmar Beier1,3, Tine Rasmussen2, Sune Munthe2,4, Bjarne Winther Kristensen2, Christoph Patrick Beier1,3

1Department of Neurology, Odense University Hospital and Institute of Clinical Research, University of Southern Denmark, Sdr. Boulevard 29, 5000 Odense, Denmark; 2Department of Pathology, Odense University Hospital and Institute of Clinical Research, University of Southern Denmark, J. B. Winsløws Vej 15, 5000 Odense, Denmark; 3Department of Neurology, RWTH Aachen, Pauwelsstrasse 30, 52074 Aachen, Germany; 4Department of Neurosurgery, University Hospital Odense and Clinical Institute, University of Southern Denmark, Sdr. Boulevard 29, 5000 Odense, Denmark

Received March 10, 2016; Accepted May 25, 2016; Epub July 1, 2016; Published July 15, 2016

Abstract: The histologically defined tumor entity glioblastoma (GBM) comprises different molecular subgroups. On a transcriptional level, at least two subgroups were uniformly identified: proneural and mesenchymal GBM. Treatment of GBM with radiotherapy and anti-angiogenic drugs may trigger molecular subtype transformation with proneural tumors shifting towards a mesenchymal phenotype. Because both treatment modalities also result in dramatic changes of the metabolic microenvironment of GBM, we hypothesized if the major determinants in local microen-vironment-hypoxia and (lactate) acidosis-may induce consistent changes in the phenotype of primary glioblastoma stem cell lines in vitro. A panel of 5 different proneural GBM stem cell lines (GSC) lines was systematically chal-lenged for 7 days with acidosis (pH 5.5-7.4), sodium lactate (0-60 mM), lactic acid (0-15 mM), or with 72 h hypoxia (2% oxygen). Acidosis significantly improved survival but inhibited proliferation of GSC lines. Sodium lactate inhibited proliferation without affecting survival. Neither acidosis nor sodium lactate did consistently change growth pattern or stem cell marker expression. Hypoxia and lactate induced a more proneural-like phenotype with increased Olig2 and CD133 expression in 1 GSC line, but did not consistently alter marker expression in the other 4 GSC lines. None of the 5 GSC lines investigated showed changes indicative for a proneural-to-mesenchymal transformation towards a more malignant phenotype. Taken together, our data does not exclude the possibility that metabolic factors like lactate acidosis may induce a more mesenchymal phenotype. Nevertheless, data suggest that therapy-induced metabolic changes in the microenvironment are likely not of crucial importance to our understanding of mesenchy-mal transformation after radio-and anti-angiogenic therapy for most GBM.

Keywords: Metabolic microenvironment, glioma stem cells, CD133, CD44, hypoxia, lactate

Introduction

Glioblastoma (GBM) is the most common pri-mary malignant brain tumor with an average survival of affected patients of about 15 months [1]. All GBM share same histological features but differ substantially on a molecular level. Various attempts have been made to divide GBM into different groups based on their profile on mRNA, protein, genome, and stem cell level [2-7]. Although all studies varied with respect to their results and the number of sub-groups identified, all studies agreed on the exi-

stence of two major groups: a (pro-) neural and a mesenchymal phenotype, which account for the majority of all GBM. These two groups were initially named because of the expression of marker genes resembling different stages of neurogenesis [3]. The subtypes differ substan-tially with respect to tumor cell properties (e.g. differences in WNT-, TGF-beta, hippo-signaling) but possibly also with respect to local microen-vironment, amongst others extend of immune infiltration [8, 9]. It remains controversial if both subgroups have prognostic significance for patients treated with standard therapy [3, 7,

http://www.ijcep.com

Metabolic challenge of glioblastoma stem cell lines

6725 Int J Clin Exp Pathol 2016;9(7):6724-6737

10]. Differences of both groups in response to immunotherapy have, though, been observed [11].

GBM are characterized by massive angiogene-sis, and new antiangiogenic therapies raised hopes to improve the patients’ outcome. Glioma-induced angiogenesis can be visualized using MRI as tumor core with pathological blood vessels and contrast enhancement [12-15]. The most thoroughly investigated sub-stance-bevacizumab (BEV)-induces often a dramatic response of the contrast-enhancing tumor in MRI and substantially improves the progression free survival (PFS) of patients irre-spective of the molecular subtype [13, 16, 17]. However, BEV failed in recent clinical studies to improve overall survival [16]. The combination of an impressive improvement of PFS without improvement in overall survival suggests a malignant transformation of the tumor under therapy. This malignant transformation seems to depend on the molecular subtype: a retro-spective analysis of the AVAGlio [16] data could recently show, that tumors with a mesenchymal subtype did not benefit from BEV treatment, while patients with proneural GBM did [17].

If and how a transformation of the molecular phenotype influences these differences remains controversial. Piao et al. showed data suggesting that treatment failure after BEV may be caused by a proneural-to-mesenchymal transformation [18]. A molecular analysis of the molecular phenotypes by DeLay et al. before and after therapy showed transition of the phe-notype after BEV treatment in 8 of 13 tumors investigated [19]. However, they found a BEV induced a transformation in both directions (i.e. proneural-to-mesenchymal and vice versa). A transformation of the molecular phenotypes was also described for recurrent gliomas after radiotherapy by Phillips et al.: 8 out of 26 paired samples investigated changed the molecular phenotype mainly from a proneural to a mesen-chymal profile [3]. In conclusion, these and other reports [20] support the hypothesis that a transformation of the molecular subtype may contribute to treatment failure.

Treatment with BEV and radiotherapy results in relevant changes of the important determi-nants of the metabolic microenvironment [21]. Impairing and destroying tumor angiogenesis results in tissue hypoxia, increased glycoly-sis, and lactate acidosis. Conversely, metabolic

changes have been postulated determinants of treatment failure/success under BEV treat-ment [22]. It is therefore possible that these factors may be the crucial trigger for this trans-formation of the molecular phenotype.

GSC lines became an accepted and widely used in vitro model to investigate GBM biology. GSC lines are more favorable compared to serum-cultured cell lines because they are genetically stabile and closely resemble the ini-tial tumor also after many passages in vitro [23]. Like GBM, GSC lines derived from tumors are heterogeneous, and, as in the parental tumors, GSC lines with a proneural and a mesenchymal transcription profile have been described [6, 8, 24]. Both types of GSC differ with respect to marker expression, although, to date, no unambiguous marker for the two major subtypes has been identified. Recently, several studies independently suggested CD133 and CD44 as markers for proneural and mesenchy-mal GSC [6, 25, 26]. In vivo, Olig2 and DLL3 have been proposed to be indicative for pro-neural and YKL-40 for mesenchymal GBM [3, 8]. In vitro, these GSC lines may differ with pro-neural GSC showing a neurosphere-like growth pattern and mesenchymal GSC having a more adherent growth pattern [24, 27]. However, the analysis of the growth pattern is not without ambiguity because culture conditions (e.g. con-tinuous selection of spheres, laminin coating, or frequent use of trypsin) change the growth pattern without affecting the molecular subtype.

The molecular mechanisms of transformation from a proneural to a more mesenchymal phe-notype were recently investigated using GSC lines. These studies showed that alterations of TGF-beta [8, 28-30], TNF-alpha/NFkB, and MET-signaling [31, 32] were of importance especially for proneural-to-mesenchymal trans-formation. The putative mechanisms of the mesenchymal-to-proneural transformation after BEV treatment are to date unknown.

In contrast to in vivo studies, the study of GSC allows to investigate changes in the tumor cells only. This is relevant because invading neural stem cells or mesenchymal cells may substan-tially modulate the molecular phenotype of tumor samples without a relevant change of bona fide tumor cells. We therefore used a panel of GSC lines to address the question if metabolic challenge induces a proneural-to-



mesenchymal transformation in proneural GSC lines to substantiate or to falsify the hypothesis that metabolic changes may induce a mesen-chymal transformation of GBM cells.

Materials and methods

Cell culture

All GSC lines were derived from primary tumor tissue dissociated direct after operation as described previously [24, 33]. All primary tumors were IDH wildtype GBM diagnosed according to the WHO classification [34]. The generation of GSC was approved by the local Ethic Committees of Southern Denmark and at the University Hospital Regensburg. GSC lines were cultured in DMEM-F12 supplemented with 20 ng/ml EGF, 20 ng/ml bFGF, B27 (1:50), and antibiotics as described previously [6, 24].

Microarray analysis

Microarray analyses were previously described in [6, 24, 35]. In brief, total RNA was isolated using the Qiagen RNeasy kit. The generation of labeled cRNA and its hybridization to U133 Plus 2 GeneChip arrays (Affymetrix) were done and scanned at the Kompetenzzentrum für Fluoreszente Bioanalytik (KFB; Regensburg, Germany) and the Department of Clinical Genetics (Odense University Hospital, Odense, Denmark) according to standard procedures [24]. Analyses were performed using BRB-Array Tools developed by Dr. Richard Simon and BRB-Array Tools Development Team. Additional microarrays of established GSC lines and pub-lished by Gunter et al. [27] and Lottaz et al. [6] were used for analysis. Arrays were normalized using RMA. Supervised clustering was per-formed using the 24 genes signature published by Lottaz et al. [6].

Metabolic challenge

Acidosis: To induce acidosis, DMEM-F12 medi-um was acidified using 37% HCl until the pH given was reached. For the experiments, the growth factors 20 ng/ml EGF, 20 ng/ml bFGF, B27 (1:50), and antibiotics were added.

Lactate acidosis: For lactate acidosis, sodium lactate was treated with equimolar concentra-tions of HCl resulting in lactic acid and NaCl. Lactic acid was added to standard culture medium to a final concentration of 5 and 15 mM.

Sodium lactate: Sodium lactate was added to standard culture medium in the concentrations given in the experiments. Due to the high con-centrations used (up to 60 mM) we chose to use glucose to as control for osmolality.

Hypoxia: GSC lines were cultured as small spheres in T25 flasks. They were cultured under moderate hypoxia (2% O, 5% CO2, 94% N2) in a Sanyo O2/CO2 incubator (Sanyo Electric Co. Ltd., Japan) for 24 h and 72 h. After the treat-ment the cells were re-seeded as single cells in 6-well plates in pairs (20.000 in each well in the case of R28 24 h and 50.000 for the rest). After 3 and 7 days pictures were taken and the media was changed.

Flow cytometry

Flow cytometry was performed according stan-dard protocols. At the time points indicated, cells were trypsinized and harvested as single cells for cytometric analysis. Each sample/treatment was split into 3 FACS samples for Isotype control (mIgG2b-PE, 1:100), CD133CD- 133/2-PE (1:100, Miltenyi Biotech), and CD44-FITC (1:100, BD), and PI staining (Sigma Aldrich). 10.000 single cells were counted for each sam-ple in a FACS Canto analyzer and the data were analyzed using FACS Diva software.

Western blot

Western blot was performed according to stan-dard protocols. At the time points indicated, cells were harvested and total protein lysates were made using Cell Lysis Reagent (Sigma) containing both Protease Inhibitor Cocktail Set I and Phosphatase Inhibitor Cocktail Set II (Calbiochem). The extracts were separated by SDS-PAGE (4-12% Bis-Tris-gel, Novex) and transferred to a 0.45 µm PVDF membrane (Invitrogen). The Novex Sharp Pre-stained pro-tein standard was used to assess the protein size. The primary antibodies used were anti-YKL-40 (Quidel, 1:100), anti-Olig2 (Bio Labs 1:100), and β-actin (N-term, AC-15 Sigma 1:10.000). The secondary antibodies were rab-bit IgG HRP-conjugated (Olig2, YKL-40) and mouse IgG HRP-conjugated (β-actin).

Results

Characterization of the GSC lines

All GSC lines used were previously character-ized in detail [6, 24, 35]. We recently could show that GSC can be classified as “proneural-



like” and “mesenchymal-like” using a 24-gene signature [6]. Supervised clustering of mRNA expression profile of the 5 GSC lines used and a panel of 20 previously characterized GSC lines showed that all GSC lines were classified as “proneural-like” based on these signature genes (Figure 1). With respect to marker expression, 4 out of the 5 GSC lines expressed the surface epitope CD133, and all GCS lines expressioned the transcription factor Olig2, which are both marker proteins associated with

Figure 1. Classification of previously published GSC lines by Gunther et al. [27], Lottaz et al. [6], and the GSC lines used (marked in yellow) according to the “Lottaz signature”.

Table 1. Marker profileCD133 Olig2 CD44 YKL-40

R28 +++ ++ - ++T78 ++ +++ ++ -T86 + ++ ++ +T87 - +++ + +T111 (+) + ++ +++++>50% positive cells/very strong signal WB; ++10-50% positive cells/strong signal WB; +1-10% positive cells/weak signal WB; (+) 0.1-1% positive cells/very weak signal WB.



“proneural-like” GSC [6, 25]. However, 4 out of 5 GCS lines also expressed CD44, and 2 out of 5 were positive for YKL-40, which are both putative markers for “mesenchymal-like GSC”. Table 1 gives and overview over the marker pro-file of the GSC lines investigated.

Effects of low pH on 5 GSC lines

Low pH (due to lactate acidosis) and low oxygen pressure/tissue hypoxia are the major compo-nents of the cellular metabolic microenviron-ment and are closely related. We therefore

Figure 2. Effects of acidosis on proliferation, survival, CD133/CD44 expression, growth pattern. (A) Cell number; (B) Proportion of viable cells; (C) Proportion of CD133 expressing cells; and (D) proportion of CD44 expressing cells after treatment with ph 6.5 and pH 5.5 for 7 days are given; (E) Representative pictures of the growth pattern of GSC lines R28 are shown.



investigated the effects of chronic expose with major metabolic determinants-acidosis, lactate concentration, lactate acidosis, and hypoxia-on

survival, proliferation, and in vitro phenotype of a panel of 5 different GSC lines. To differentiate the effects of lactate acidosis and low pH, we

Figure 3. Effects of sodium lactate on proliferation, survival, CD133/CD44 expression, growth pattern. (A) Cell num-ber; (B) Proportion of viable cells; (C) Proportion of CD133 expressing cells; and (D) proportion of CD44 expressing cells after treatment with 30 mM Na-lactate and 60 mM sodium lactate for 7 days are given. Medium supplemented with 60 mM glucose was chosen as control condition; (E) Representative pictures of the growth pattern of GSC lines R28 are shown.



Figure 4. Effects of lactate acidosis on prolif-eration, survival, CD133/CD44 expression, growth pattern, Olig2/YKL-40 expression. (A)Cell number; (B) Proportion of viable cells; (C) Proportion of CD133 expressing cells; and (D) proportion of CD44 expressing cells after treatment with 5 mM and 15 mM lactate for 7 days are given; (E) Representative pictures of the growth pattern of GSC lines R28 are shown; (F) Western blot for Olig2 and YKL-40 expression of the panel of GSC lines treated with 15 mM lactate for 7 days (C = control, L = treatment with 15 mM lactate).



challenged the 5 GSC lines using standard serum free medium acidified using HCl to pH

6.5 and 5.5 for 7 days. Acidification significant-ly and dose-dependently reduced the prolifera-

Figure 5. Effects of acute hypoxia on proliferation, survival, CD133/CD44 expression, growth pattern. (A) Cell num-ber; (B) Proportion of viable cells; (C) Proportion of CD133 expressing cells; and (D) proportion of CD44 expressing cells after treatment with 2% O2 for 72 h days are given; (E) Representative pictures of the growth pattern of GSC lines R28 are shown; (F) Western blot for Olig2 and YKL-40 expression of the panel of GSC lines treated with hypoxia (2% oxygen) for 72 h (c = control, H = hypoxia).



tion of GSC lines (Figure 2A). Remarkably, acidi-fication also increased survival as determined by a significantly increased proportion of viable cells as determined by PI staining (Figure 2B). Despite these marked effect, acidification did neither significantly changed expression of the surface marker of CD133 and CD44 (Figure 2C, 2D), nor the growth pattern of the GSC lines (Figure 2E).

Effects of sodium lactate on 5 GSC lines

Lactate acidosis is a major contributor to the metabolic microenvironment especially in tumors. To dissect the effects of lactate alone and lactate acidosis, we treated the panel of 5 GSC with increasing doses of sodium lactate. As expected, sodium lactate significantly inhib-ited proliferation of GSC lines (Figure 3A) with-out consistently affecting survival of the cells (Figure 3B). With respect to stem cell marker expression, sodium lactate did not induce sig-nificant and consistent changes in the expres-sion of CD133 and CD44 (Figure 3C, 3D). Further, the growth pattern of the GSC was unaffected (Figure 3E).

Effects of lactate acidosis on GSC lines

Due to the “Warburg effect”, lactate acidosis represents a hallmark of cancer and is also present in GBM [36-38]. In vitro, increasing doses of lactate acidosis did not elicit homoge-neous effects in the 5 GSC lines screened. Proliferation was impaired in two GSC lines (T86, T78), which were also particular sensitive to lactate and acidosis induced growth inhibi-tion (Figure 4A). The concentrations used for 1 week (5 mM and 15 mM lactate) did not increase proliferation of a single GSC line and did not consistently affect survival (Figure 4A, 4B). With respect to stem cell marker expres-sion, lactate acidosis induced CD133 expres-sion in the GSC line T78. The other GSC lines did not show a relevant change in proneural (CD133, Figure 4C) and mesenchymal (CD44, Figure 4D) marker expression. Growth pattern was slightly altered in one GSC line (T111) towards a more adherent growth pattern (Figure 4E). Because of the slight effects seen in these experiments, we performed confirma-tory experiments investigating the expression of YKL-40 and Olig2 by Western blot. Both are proposed marker proteins for mesenchymal and proneural-like GSC lines. Lactate acidosis

induced an increased Olig2 expression in the GSC line T78, which also showed an increase of CD133 expression, and in the GSC line R28, which did not show a relevant change of CD133 expression. The expression level of YKL-40 was not substantially altered by lactate acidosis.

Short-term and long-term effects of hypoxia on GSC lines

Hypoxia is the other determinant of the meta-bolic microenvironment and the effects of hypoxia on GSC lines have been extensively studied [35, 39, 40]. For this experiment, GSC lines were incubated with 2% oxygen for 72 h and analyzed direct after exposure and 7 days after re-plating of treated cells. There were no consistent changes in all 5 GSC lines. Hypoxia foster proliferation in GSC lines R28, T78, and T86 as compared to control conditions, the proliferation of GSC T87 was impaired under hypoxia but cells recovered rapidly (Figures 5A, 5B and 6A, 6B). Notably, 2% hypoxia did not induce cell death but seemed to have protec-tive properties (Figure 5B). Only the phenotype of GSC line T111 was substantially altered: 72 h hypoxia slightly reduced the number of cells but the cells kept proliferating substantially slower than control cells (Figures 5A, 6A). The reduced proliferation of GSC lines T111 was paralleled by a decrease of CD133 and an increased expression of CD44 but without a change of the growth properties, loss of Olig2 expression or increase of YKL-40 expression (Figures 5C-F, 6C-E). In the other 4 GSC lines, hypoxia slightly increased the expression of CD133 in two GSC lines. Because of the effects seen in the GSC lines T111, we chose to study the expression of Olig2 and YKL-40 in all GSC lines upon hypoxia. Hypoxia substantially induced Olig2 expression in the GSC line T78 and slightly increased Olig2 expression in R28 [41]. With respect to YKL-40, an increase of the expression level was seen in R28, T78 showed minimal reduction (Figure 5F).

Discussion

A substantial part of GBM patients showed shrinkage of the contrast-enhancing part of the tumor upon anti-angiogenic therapy. Un- fortunately, this did not translate into a longer survival and may change the tumors’ growth pattern [42]. These clinical phenomena, also seen with other anti-angiogenic treatments,



may suggest a fundamental change of tumor properties. Piao et al. postulated that a proneu-ral-to-mesenchymal transformation causes the acquisition of a more mesenchymal phenotype and thereby contributes to the resistance against anti-angiogenic therapy [18, 43]. Given that the mesenchymal phenotype may also be associated with a worse prognosis [3], this mesenchymal transformation may explain the more aggressive behavior of recurrent GBM especially after anti-angiogenic treatment. Data by DeLay et al. could not support the con-cept proposed by Piao et al. but the mesenchy-mal signature of a tumor does not necessarily only depend on tumor cells but also on cells within the tumor stroma, like endothelial cells.

In vitro, we could anecdotally observe a trans-formation of the expression profile of long-term GSC lines (unpublished observation). We there-fore aimed at establishing an in vitro model allowing investigating how metabolic factors modulate the properties of proneural GSC lines. We screened at total of 8 GSC lines, 5 GSC lines were investigated extensively. We could not observe any evidence for a transfor-mation into a more mesenchymal subtype. Thus, we would like to contribute to the ongoing discussion on the relevance and pathomecha-nism of the postulated mesenchymal transfor-mation, by providing evidence that chronic changes in the metabolic microenvironment were not able to induce a relevant mesenchy-mal transformation in our panel of proneural GSC lines. Quite contrary to this hypothesis, we found a GSC line (T78), which showed a more proneural-like phenotype with induction of Olig2 and CD133 expression. We did not go into detail if this increase of marker expression actually corresponded to induction of “stem-ness” or a more pronounced proneural pheno-type. Either way there was no indication for a more mesenchymal phenotype. Furthermore, we did not make microarray analysis after met-abolic challenge of the GSC lines. The analysis of the cell cultures did neither indicate a change of the growth pattern, nor a remarkable increase of migrating cells. Thus, microrarray analysis would not have contributed to our

understanding of the lack of a mesenchymal transformation.

Of course, our experiment does not exclude the possibility that a proneural-to-mesenchymal transformation does exist. We could have used more GSC lines and we did not test all possible conditions: We neither used extreme condi-tions, nor did we investigate the effects of very long treatment (>1 week). Even if the investiga-tion of 20 GSC lines would have shown sub-stantial effects in 1-2 GSC lines under extreme conditions, our major conclusion would have remained unchanged: A substantial proneural-to-mesenchymal transformation triggered by chronic physiological metabolic changes is not common and therefore likely not relevant for the vast majority of patients.

In summary, our experiments did not provide any evidence in support of the concept post- ulating that the proneural-to-mesenchymal transformation induced by changes of the local metabolic microenviroment is common and relevant.

Acknowledgements

This work was supported by the free research grants of the Odense University Hospital to DB and CPB.

Disclosure of conflict of interest

None.

Address correspondence to: Dr. Christoph Patrick Beier, Department of Neurology, Odense University Hospital and Institute of Clinical Research, Univer- sity of Southern Denmark, Sdr. Boulevard 29, 5000 Odense, Denmark. Tel: +45 6541 1943; Fax: +45 6541 3389; E-mail: [email protected]

References

[1] Stupp R, Mason WP, van den Bent MJ, Weller M, Fisher B, Taphoorn MJ, Belanger K, Brandes AA, Marosi C, Bogdahn U, Curschmann J, Janzer RC, Ludwin SK, Gorlia T, Allgeier A, Lacombe D, Cairncross JG, Eisenhauer E and Mirimanoff RO. Radiotherapy plus concomi-

Figure 6. Long-term effects of hypoxia on proliferation, survival, CD133/CD44 expression, growth pattern. (A) Cell number, (B) Proportion of viable cells, (C) Proportion of CD133 expressing cells, and (D) proportion of CD44 express-ing cells 7 days after treatment 2% hypoxia for 72 h are given. (E) Representative pictures of the growth pattern of GSC lines R28 7 days after hypoxia are shown.



tant and adjuvant temozolomide for glioblas-toma. N Engl J Med 2005; 352: 987-996.

[2] Huse JT, Phillips HS and Brennan CW. Mole- cular subclassification of diffuse gliomas: see-ing order in the chaos. Glia 2011; 59: 1190-1199.

[3] Phillips HS, Kharbanda S, Chen R, Forrest WF, Soriano RH, Wu TD, Misra A, Nigro JM, Colman H, Soroceanu L, Williams PM, Modrusan Z, Feuerstein BG and Aldape K. Molecular sub-classes of high-grade glioma predict progno-sis, delineate a pattern of disease progression, and resemble stages in neurogenesis. Cancer Cell 2006; 9: 157-173.

[4] Li A, Walling J, Ahn S, Kotliarov Y, Su Q, Quezado M, Oberholtzer JC, Park J, Zenklusen JC and Fine HA. Unsupervised analysis of transcrip-tomic profiles reveals six glioma subtypes. Cancer Res 2009; 69: 2091-2099.

[5] Brennan C, Momota H, Hambardzumyan D, Ozawa T, Tandon A, Pedraza A and Holland E. Glioblastoma subclasses can be defined by ac-tivity among signal transduction pathways and associated genomic alterations. PLoS One 2009; 4: e7752.

[6] Lottaz C, Beier D, Meyer K, Kumar P, Hermann A, Schwarz J, Junker M, Oefner PJ, Bogdahn U, Wischhusen J, Spang R, Storch A and Beier CP. Transcriptional profiles of CD133+ and CD133-glioblastoma-derived cancer stem cell lines suggest different cells of origin. Cancer Res 2010; 70: 2030-2040.

[7] Verhaak RG, Hoadley KA, Purdom E, Wang V, Qi Y, Wilkerson MD, Miller CR, Ding L, Golub T, Mesirov JP, Alexe G, Lawrence M, O’Kelly M, Tamayo P, Weir BA, Gabriel S, Winckler W, Gupta S, Jakkula L, Feiler HS, Hodgson JG, James CD, Sarkaria JN, Brennan C, Kahn A, Spellman PT, Wilson RK, Speed TP, Gray JW, Meyerson M, Getz G, Perou CM, Hayes DN; Cancer Genome Atlas Research Network. Integrated genomic analysis identifies clinical-ly relevant subtypes of glioblastoma character-ized by abnormalities in PDGFRA, IDH1, EGFR, and NF1. Cancer Cell 2010; 17: 98-110.

[8] Beier CP, Kumar P, Meyer K, Leukel P, Bruttel V, Aschenbrenner I, Riemenschneider MJ, Fragoulis A, Rummele P, Lamszus K, Schulz JB, Weis J, Bogdahn U, Wischhusen J, Hau P, Spang R and Beier D. The cancer stem cell subtype determines immune infiltration of glio-blastoma. Stem Cells Dev 2012; 21: 2753-2761.

[9] Bhat KP, Balasubramaniyan V, Vaillant B, Ezhilarasan R, Hummelink K, Hollingsworth F, Wani K, Heathcock L, James JD, Goodman LD, Conroy S, Long L, Lelic N, Wang S, Gumin J, Raj D, Kodama Y, Raghunathan A, Olar A, Joshi K, Pelloski CE, Heimberger A, Kim SH, Cahill DP, Rao G, Den Dunnen WF, Boddeke HW, Phillips

HS, Nakano I, Lang FF, Colman H, Sulman EP and Aldape K. Mesenchymal differentiation mediated by NF-kappaB promotes radiation re-sistance in glioblastoma. Cancer Cell 2013; 24: 331-346.

[10] Murat A, Migliavacca E, Gorlia T, Lambiv WL, Shay T, Hamou MF, de Tribolet N, Regli L, Wick W, Kouwenhoven MC, Hainfellner JA, Heppner FL, Dietrich PY, Zimmer Y, Cairncross JG, Janzer RC, Domany E, Delorenzi M, Stupp R and Hegi ME. Stem cell-related “self-renewal” signature and high epidermal growth factor receptor ex-pression associated with resistance to con-comitant chemoradiotherapy in glioblastoma. J Clin Oncol 2008; 26: 3015-3024.

[11] Prins RM, Soto H, Konkankit V, Odesa SK, Eskin A, Yong WH, Nelson SF and Liau LM. Gene expression profile correlates with T-cell infiltration and relative survival in glioblastoma patients vaccinated with dendritic cell immu-notherapy. Clin Cancer Res 2011; 17: 1603-1615.

[12] Kreisl TN, Kim L, Moore K, Duic P, Royce C, Stroud I, Garren N, Mackey M, Butman JA, Camphausen K, Park J, Albert PS and Fine HA. Phase II Trial of Single-Agent Bevacizumab Followed by Bevacizumab Plus Irinotecan at Tumor Progression in Recurrent Glioblastoma. J Clin Oncol 2009; 27: 740-745.

[13] Vredenburgh JJ, Desjardins A, Herndon JE 2nd, Marcello J, Reardon DA, Quinn JA, Rich JN, Sathornsumetee S, Gururangan S, Sampson J, Wagner M, Bailey L, Bigner DD, Friedman AH and Friedman HS. Bevacizumab plus irinote-can in recurrent glioblastoma multiforme. J Clin Oncol 2007; 25: 4722-4729.

[14] Stupp R, Hegi ME, Neyns B, Goldbrunner R, Schlegel U, Clement PM, Grabenbauer GG, Ochsenbein AF, Simon M, Dietrich PY, Pietsch T, Hicking C, Tonn JC, Diserens AC, Pica A, Hermisson M, Krueger S, Picard M and Weller M. Phase I/IIa study of cilengitide and temo-zolomide with concomitant radiotherapy fol-lowed by cilengitide and temozolomide mainte-nance therapy in patients with newly diag-nosed glioblastoma. J Clin Oncol 2010; 28: 2712-2718.

[15] Wick W, Weller M, Weiler M, Batchelor T, Yung AW and Platten M. Pathway inhibition: emerg-ing molecular targets for treating glioblastoma. Neuro Oncol 2011; 13: 566-579.

[16] Chinot OL, de La Motte Rouge T, Moore N, Zeaiter A, Das A, Phillips H, Modrusan Z and Cloughesy T. AVAglio: Phase 3 trial of bevaci-zumab plus temozolomide and radiotherapy in newly diagnosed glioblastoma multiforme. Adv Ther 2011; 28: 334-340.

[17] Sandmann T, Bourgon R, Garcia J, Li C, Cloughesy T, Chinot OL, Wick W, Nishikawa R, Mason W, Henriksson R, Saran F, Lai A, Moore



N, Kharbanda S, Peale F, Hegde P, Abrey LE, Phillips HS and Bais C. Patients With Proneural Glioblastoma May Derive Overall Survival Benefit From the Addition of Bevacizumab to First-Line Radiotherapy and Temozolomide: Retrospective Analysis of the AVAglio Trial. J Clin Oncol 2015; 33: 2735-2744.

[18] Piao Y, Liang J, Holmes L, Henry V, Sulman E and de Groot JF. Acquired resistance to anti-VEGF therapy in glioblastoma is associated with a mesenchymal transition. Clin Cancer Res 2013; 19: 4392-4403.

[19] DeLay M, Jahangiri A, Carbonell WS, Hu YL, Tsao S, Tom MW, Paquette J, Tokuyasu TA and Aghi MK. Microarray analysis verifies two dis-tinct phenotypes of glioblastomas resistant to antiangiogenic therapy. Clin Cancer Res 2012; 18: 2930-2942.

[20] Halliday J, Helmy K, Pattwell SS, Pitter KL, LaPlant Q, Ozawa T and Holland EC. In vivo ra-diation response of proneural glioma charac-terized by protective p53 transcriptional pro-gram and proneural-mesenchymal shift. Proc Natl Acad Sci U S A 2014; 111: 5248-5253.

[21] Keunen O, Johansson M, Oudin A, Sanzey M, Rahim SA, Fack F, Thorsen F, Taxt T, Bartos M, Jirik R, Miletic H, Wang J, Stieber D, Stuhr L, Moen I, Rygh CB, Bjerkvig R and Niclou SP. Anti-VEGF treatment reduces blood supply and increases tumor cell invasion in glioblastoma. Proc Natl Acad Sci U S A 2011; 108: 3749-3754.

[22] Sathornsumetee S, Cao Y, Marcello JE, Herndon JE 2nd, McLendon RE, Desjardins A, Friedman HS, Dewhirst MW, Vredenburgh JJ, Rich JN. Tumor angiogenic and hypoxic profiles predict radiographic response and survival in malignant astrocytoma patients treated with bevacizumab and irinotecan. J Clin Oncol 2008; 26: 271-278.

[23] Lee J, Kotliarova S, Kotliarov Y, Li A, Su Q, Donin NM, Pastorino S, Purow BW, Christopher N, Zhang W, Park JK and Fine HA. Tumor stem cells derived from glioblastomas cultured in bFGF and EGF more closely mirror the pheno-type and genotype of primary tumors than do serum-cultured cell lines. Cancer Cell 2006; 9: 391-403.

[24] Beier D, Hau P, Proescholdt M, Lohmeier A, Wischhusen J, Oefner PJ, Aigner L, Brawanski A, Bogdahn U and Beier CP. CD133+ and CD133-glioblastoma-derived cancer stem cells show differential growth characteristics and molecular profiles. Cancer Res 2007; 67: 4010-4015.

[25] Brown DV, Daniel PM, D’Abaco GM, Gogos A, Ng W, Morokoff AP and Mantamadiotis T. Coexpression analysis of CD133 and CD44 identifies proneural and mesenchymal sub-

types of glioblastoma multiforme. Oncotarget 2015; 6: 6267-6280.

[26] Mao P, Joshi K, Li J, Kim SH, Li P, Santana-Santos L, Luthra S, Chandran UR, Benos PV, Smith L, Wang M, Hu B, Cheng SY, Sobol RW and Nakano I. Mesenchymal glioma stem cells are maintained by activated glycolytic metabo-lism involving aldehyde dehydrogenase 1A3. Proc Natl Acad Sci U S A 2013; 110: 8644-8649.

[27] Gunther HS, Schmidt NO, Phillips HS, Kemming D, Kharbanda S, Soriano R, Modrusan Z, Meissner H, Westphal M and Lamszus K. Glioblastoma-derived stem cell-enriched cul-tures form distinct subgroups according to molecular and phenotypic criteria. Oncogene 2008; 27: 2897-2909.

[28] Joseph JV, Conroy S, Pavlov K, Sontakke P, Tomar T, Eggens-Meijer E, Balasubramaniyan V, Wagemakers M, den Dunnen WFA and Kruyt FAE. Hypoxia enhances migration and invasion in glioblastoma by promoting a mesenchymal shift mediated by the HIF1α-ZEB1 axis. Cancer Lett 2015; 359: 107-116.

[29] Joseph J, Conroy S, Tomar T, Eggens-Meijer E, Bhat K, Copray S, Walenkamp A, Boddeke E, Balasubramanyian V and Wagemakers M. TGF-β is an inducer of ZEB1-dependent mes-enchymal transdifferentiation in glioblastoma that is associated with tumor invasion. Cell Death Dis 2014; 5: e1443.

[30] Chandran UR, Luthra S, Santana-Santos L, Mao P, Kim SH, Minata M, Li J, Benos PV, DeWang M, Hu B, Cheng SY, Nakano I and Sobol RW. Gene expression profiling distin-guishes proneural glioma stem cells from mes-enchymal glioma stem cells. Genom Data 2015; 5: 333-336.

[31] Jahangiri A, De Lay M, Miller LM, Carbonell WS, Hu YL, Lu K, Tom MW, Paquette J, Tokuyasu TA, Tsao S, Marshall R, Perry A, Bjorgan KM, Chaumeil MM, Ronen SM, Bergers G and Aghi MK. Gene expression profile identifies tyrosine kinase c-Met as a targetable mediator of anti-angiogenic therapy resistance. Clin Cancer Res 2013; 19: 1773-1783.

[32] De Bacco F, Casanova E, Medico E, Pellegatta S, Orzan F, Albano R, Luraghi P, Reato G, D'Ambrosio A, Porrati P, Patane M, Maderna E, Pollo B, Comoglio PM, Finocchiaro G and Boccaccio C. The MET oncogene is a functional marker of a glioblastoma stem cell subtype. Cancer Res 2012; 72: 4537-4550.

[33] Jensen SS, Aaberg-Jessen C, Andersen C, Schroder HD and Kristensen BW. Glioma spheroids obtained via ultrasonic aspiration are viable and express stem cell markers: a new tissue resource for glioma research. Neurosurgery 2013; 73: 868-886; discussion 886.



[34] Louis DN, Ohgaki H, Wiestler OD, Cavenee WK, Burger PC, Jouvet A, Scheithauer BW and Kleihues P. The 2007 WHO classification of tu-mours of the central nervous system. Acta Neuropathol 2007; 114: 97-109.

[35] Kolenda J, Jensen SS, Aaberg-Jessen C, Christensen K, Andersen C, Brunner N and Kristensen BW. Effects of hypoxia on expres-sion of a panel of stem cell and chemoresis-tance markers in glioblastoma-derived spher-oids. J Neurooncol 2011; 103: 43-58.

[36] Hanahan D and Weinberg Robert A. Hallmarks of Cancer: The Next Generation. Cell 2011; 144: 646-674.

[37] Hirschhaeuser F, Sattler UGA and Mueller-Klieser W. Lactate: A Metabolic Key Player in Cancer. Cancer Res 2011; 71: 6921-6925.

[38] Herholz K, Heindel W, Luyten PR, denHollander JA, Pietrzyk U, Voges J, Kugel H, Friedmann G and Heiss WD. In vivo imaging of glucose con-sumption and lactate concentration in human gliomas. Ann Neurol 1992; 31: 319-327.

[39] Seidel S, Garvalov BK, Wirta V, von Stechow L, Schanzer A, Meletis K, Wolter M, Sommerlad D, Henze AT, Nister M, Reifenberger G, Lunde- berg J, Frisen J and Acker T. A hypoxic niche regulates glioblastoma stem cells through hy-poxia inducible factor 2 alpha. Brain 2010; 133: 983-995.

[40] Heddleston JM, Li Z, Lathia JD, Bao S, Hjel- meland AB and Rich JN. Hypoxia inducible fac-tors in cancer stem cells. Br J Cancer 2010; 102: 789-795.

[41] Hjelmeland AB, Wu Q, Heddleston JM, Choud- hary GS, MacSwords J, Lathia JD, McLendon R, Lindner D, Sloan A and Rich JN. Acidic stress promotes a glioma stem cell phenotype. Cell Death Differ 2011; 18: 829-840.

[42] Schaub C, Greschus S, Seifert M, Waha A, Blasius E, Rasch K, Landwehr C, Mack F, Schafer N, Stuplich M, Kebir S, Vilz B, Scheffler B, Bostrom J, Simon M, Urbach H, Glas M and Herrlinger U. FLAIR-only progression in bevaci-zumab-treated relapsing glioblastoma does not predict short survival. Oncology 2013; 85: 191-195.

[43] Piao Y, Liang J, Holmes L, Zurita AJ, Henry V, Heymach JV and de Groot JF. Glioblastoma re-sistance to anti-VEGF therapy is associated with myeloid cell infiltration, stem cell accumu-lation, and a mesenchymal phenotype. Neuro Oncol 2012; 14: 1379-1392.

syddansk universitet metabolic challenge of proneural ... · metabolic challenge of glioblastoma...

Documents