Accessing Key Steps of Human Tumor Progression in Vivo by Using an Avian Embryo Model

Accessing key steps of human tumor progression in vivo by using an avian embryo model

Martin Hagedorn*†‡, Sophie Javerzat*†, Delphine Gilges§, Aure´ lie Meyre*, Benjamin de Lafarge¶, Anne Eichmann¶, and Andreas Bikfalvi*

*Institut National de la Santeét de la Recherche Me´dicale E0113, Molecular Mechanisms of Angiogenesis, University Bordeaux I, Avenue des Faculte´s, 33405 Talence, France; §Institut National de la Santeét de la Recherche Me´dicale U602, Groupe Hospitalier Paul Brousse, 12 Avenue Paul Vaillant Couturier, 94807 Villejuif Cedex, France; and ¶Institut National de la Santeét de la Recherche Me´dicale U36, Colle`ge de France, 11 Place Marcelin Berthelot, 75005 Paris, France

Communicated by N. M. Le Douarin, Acade´mie des Sciences de l’Institut de France, Paris, France, December 16, 2004 (received for review November 16, 2004) Experimental in vivo tumor models are essential for comprehend- genome has now been sequenced (www.nhgri.nih.gov͞11510730), ing the dynamic process of human cancer progression, identifying new techniques allowing gene function studies in chick embryos are therapeutic targets, and evaluating antitumor drugs. However, emerging (7, 8), and a vascular marker for normal and angiogenic current rodent models are limited by high costs, long experimental chick capillaries has been identified (9). Anticipating the impact of duration, variability, restricted accessibility to the tumor, and these advances, we established a robust and highly reproducible major ethical concerns. To avoid these shortcomings, we investi- GBM progression model on the chorio-allantoic membrane gated whether tumor growth on the chick chorio-allantoic mem- (CAM), a densely vascularized, extraembryonic tissue. This model brane after human glioblastoma cell grafting would replicate allows fast and precise analysis of the principal steps of human characteristics of the human disease. Avascular tumors consistently GBM progression and response to comprehensive treatment with formed within 2 days, then progressed through vascular endothe- new anticancer drugs. Furthermore, we provide evidence that gene lial growth factor receptor 2-dependent angiogenesis, associated regulation in the course of this short-term assay predicts gene with hemorrhage, necrosis, and peritumoral edema. Blocking of expression in patient GBMs. vascular endothelial growth factor receptor 2 and platelet-derived growth factor receptor signaling pathways by using small-mole- Materials and Methods cule receptor tyrosine kinase inhibitors abrogated tumor develop- Cells and Embryos. U87 human glioma cells (American Type Culture ment. Gene regulation during the angiogenic switch was analyzed Collection) were maintained in DMEM with 10% FBS, antibiotics, by oligonucleotide microarrays. Defined sample selection for gene and L-glutamine. Fertilized chicken eggs (Gallus gallus)(EARL profiling permitted identification of regulated genes whose func- Morizeau, Dangers, France) were handled as described (10). tions are associated mainly with tumor vascularization and growth. Furthermore, expression of known tumor progression Experimental Glioma Assay. On embryonic day 10, a plastic ring was genes identified in the screen (IL-6 and cysteine-rich angiogenic placed on the CAM (10), and 3 million to 5 million U87 cells in 20 inducer 61) as well as potential regulators (lumican and F-box-only ␮l of medium were deposited after gentle laceration of surface. 6) follow similar patterns in patient glioma. The model reliably Digital photos were taken under a stereomicroscope (Nikon simulates key features of human glioma growth in a few days and SMZ800). When required, tumor size-matching was done based on thus could considerably increase the speed and efficacy of research tumor volume calculation: V ϭ 4͞3␲r3,withr ϭ 1͞2 ͌(d1 ϫ d2). on human tumor progression and preclinical drug screening. For edema quantification, experimental glioma at day 2 (D2) or control CAMs, which received cell-free medium, were fixed in vivo angiogenesis ͉ animal model alternatives ͉ glioblastoma with 4% paraformaldehyde and spread out on transparent plastic tubes. CAM surfaces were measured on cross sections, and statis- lioblastoma multiforme (GBM) brain tumors are among the tical comparison between the two groups was performed by using Gmost lethal human cancers with a mean survival time of only the Student’s t test. several months (1). Existing in vivo GBM models are based on inoculation of glioma cells into rodent brains (2) or the use of In Situ Hybridization, Histology, and Immunohistochemistry. In situ transgenic mice with modified expression of tumor suppressors or hybridization with a vascular endothelial growth factor receptor 2 oncogenes in glial cell lineages (3, 4). However, these models suffer (VEGFR-2) riboprobe was performed as described (11). For ␮ from generally poor penetrance or variable growth rate, a major histology, tumors were cut into 10- to 20- m cryosections and

obstacle for collecting clearly graded (prevascular͞vascularized) stained with hematoxylin-erythrosin-safran. Primary antibodies MEDICAL SCIENCES tumor samples exploitable in gene profiling experiments. Rodent used were anti-human vimentin (clone V9, NeoMarkers, Fremont, models are further limited by the difficulty of obtaining morpho- CA, 1:200 or AB1620, Chemicon, 1:40), rabbit anti-human desmin logical data during tumor progression, resulting in a high number (clone D33, DAKO, 1:500), mouse anti-chick tenascin C (TN-C, of animals required to obtain conclusive results, paralleled by an clone M1B4, 1:800), mouse anti-human TN-C [clone B2–813, 1:800 elevated level of animal suffering. The latter fact has been increas- (12)], and mouse anti cytochrome c (clone 6H2.B4, Pharmingen, ingly discussed and severely questioned by public opinion (5). A 1:200). Corresponding secondary antibodies were from Molecular reproducible and fast in vivo GBM model, in which both morpho- Probes (Alexa Fluor 488, 546, or 647; 1:2,000). Chick blood vessels logical and gene regulation steps of human GBM can be studied, would greatly facilitate discovery and validation of new therapeutic Abbreviations: GBM, glioblastoma multiforme; AS, astrocytoma; CAM, chorio-allantoic targets. Ideally, such a model would incorporate the widely ac- membrane; VEGFR-2, vascular endothelial growth factor receptor 2; LUM, lumican; CYR61, cepted guidelines aimed at reduction of animal number and suf- cysteine-rich angiogenic inducer 61; FBXO6, F-box-only 6; D(n), day (n); qPCR, quantitative fering, as well as the refinement and replacement of models (rule PCR; TN-C, tenascin C. of the 3Rs) (6). For decades, the avian embryo has been a model †M.H. and S.J. contributed equally to this work. organism of choice in developmental biology, whereas the field of ‡To whom correspondence should be addressed. E-mail: m.hagedorn@angio. basic cancer research has focused mainly on mouse models because u-bordeaux1.fr. of the potential given by genetic engineering. The complete chick © 2005 by The National Academy of Sciences of the USA

www.pnas.org͞cgi͞doi͞10.1073͞pnas.0408622102 PNAS ͉ February 1, 2005 ͉ vol. 102 ͉ no. 5 ͉ 1643–1648 Downloaded by guest on September 30, 2021 were visualized by using SNA-1 lectin coupled to FITC (1:200, Vector Laboratories). Fluorescent labeling was viewed by confocal microscopy (Olympus FluoView 500).

Treatment of Experimental Glioma with Receptor Tyrosine Kinase Inhibitors. Gleevec (Imatinib mesylate) and PTK787͞ZK 222584 were provided by Novartis (Basel). On D2, size-matched tumors were divided into control and treatment groups. Drugs were dissolved according to the manufacturers’ recommendations and deposited locally at 50 ␮g per egg per day. Controls received solvent of the corresponding drug. Quantification of drug effects on pericyte, tumor cell, and vessel density were determined in five to seven representative tumors per group. The contrast of red-green- blue images of equal pixel size was adjusted to 100% in PHOTOSHOP (Adobe Systems, San Jose, CA), and relative cell density was expressed as the percentage of pixels in the corresponding color channel per total number of pixels. Blood vessels were counted manually. Statistical analysis of control vs. Gleevec- or PTK787͞ZK 222584-treated tumors was carried out by using the Mann–Whitney U test.

Gene Expression Profiling. RNA was extracted from pools of 10 size-matched tumors at D2 and D4 by using Qiagen (Valencia, CA) RNeasy columns. Five micrograms of RNA was reverse-transcribed (cDNA Synthesis System Kit, Roche Diagnostics). The purified cDNA was transcribed in the presence of Cy3-UTP or Cy5-UTP (Megascript D7 kit, Ambion, Austin, TX). cRNAs were purified (High Pure RNA Tissue Kit Columns, Roche Diagnostics), frag- mented by heating, mixed (1:1), and hybridized in duplicates in 35-␮l volume on array slides (9,850 genes, Human 30k Array A, MWG Biotech, Ebersberg, Germany). Scanning and spot analysis were done according to the manufacturer’s guidelines (GMS 428 scanner and JAGUAR 2.0 software, Affymetrix, Santa Clara, CA). Only spots with reproducible Ͼ1.5-fold change were considered. A global mean normalization was applied to the data, and fold Fig. 1. Growth characteristics of experimental glioma on the CAM. (a) changes were expressed as a mean value Ϯ SEM from three During the first 2 days after implantation, a well defined, solid tumor forms, independent hybridizations. which progressively becomes vascularized the following days. (b) Tortuous angiogenic tumor capillaries are visible at the surface of a D7 tumor. (c) Cross Real-Time Quantitative PCR (qPCR). Experimental glioma and biop- section of paraformaldehyde-fixed glioma in situ with adjacent CAM at D7. sies [astrocytoma (AS) II and GBMs] from patients who gave their Areas of the tumor periphery and the center (rectangles) at higher magnifi- informed consent (Centre Hospitalier Universitaire, Bordeaux) cation show angiogenesis (Ang), necrosis (Nec), and hemorrhage (Hem). (d) were snap-frozen. RNA was purified by using RNeasy columns Histological analysis of D7 glioma reveals the presence of nucleated chick erythrocytes in tumor capillaries. (e) Glioma growth on the CAM is paralleled (Qiagen) and reversed-transcribed with SuperScript II RNase H- by the development of a large peritumoral edema (arrows, asterisk denotes Reverse Transcriptase (Invitrogen) by using oligo(dT)15 priming. tumor). CAMs close to the tumor are six times more swollen than controls. Human-specific primers for qPCR were designed and evaluated [Magnifications: a, ϫ10 (bar, 1 mm); b and c, ϫ63; d, ϫ100; and e, ϫ2.] for amplification efficiency. Primer sequences were: ␣-tubulin (NM࿝006082), 5Ј-GAGTGCATCTCCATCCACGTT-3Ј,5Ј- TAGAGCTCCCAGCAGGCATT-3Ј; IL-6,5Ј-CCAGGAGC- FBXO6, n ϭ 9 patients). For LUM (n ϭ 12 patients) and CYR61 CCAGCTATGAAC-3Ј,5Ј-CCCAGGGAGAAGGCAACTG-3Ј; (n ϭ 12 patients), values from one representative qPCR experiment cysteine-rich angiogenic inducer 61 (CYR61), 5Ј-TGAGGAGCAT- were plotted, after statistical confirmation of the reproducibility TAAGGTATTTCGAA-3Ј,5Ј-CGTGGCTGCATTAGTGTC- between experiment replicates (R2 Ͼ 0.99). CAT-3Ј; lumican (LUM), 5Ј-GATGGCAATCGCATCTCAGA- 3Ј,5Ј-TCGTTAGCAACACGTAGACATTCA-3Ј; and F-box-only Results 6(FBXO6), 5Ј-TCCTACG-GAGCCTGCATAGG-3Ј,5Ј-GCCAT- Experimental Glioma Recapitulate Hallmarks of Human GBM. U87 GCAAACATATCCTCTTCA-3Ј. Real-time PCR was carried out glioma cells deposited on the CAM form a solid, avascular tumor in a Mx3000P thermocycler (Stratagene) by using SYBR Green dye within the first day. Vascularization of the tumor steadily progresses (ABgene, Epsom, U.K.). All samples were tested in a minimum of as evidenced by biomicroscopy in vivo (Fig. 1a). At high magnifi- two independent experiments. cation, tortuous and dilated vessels at the tumor surface were clearly To confirm microarray data in experimental glioma, relative visible (Fig. 1b), and blood flow could be observed (data not changes of expression were expressed for each target gene after shown), providing direct evidence for functional tumor capillaries. normalization by the tubulin signal, according to the formula: 2⌬⌬Ct Histological analysis confirms that nucleated chick erythrocytes are with ⌬Ct ϭԽCttubulin Ϫ CttargetԽ, and ⌬⌬Ct ϭ (⌬CtD4 Ϫ⌬CtD2). present in tumor capillaries (Fig. 1d). Expression of selected genes in patients with GBM was compared Coexistence of angiogenesis, necrosis, and hemorrhage is patho- with a pool of ASII by using ⌬⌬Ct ϭ (⌬CtGBM Ϫ⌬CtASII), with gnomonic for human GBM. Transverse sections of 7-day-old ⌬CtASII calculated from the average of four to five ASIIs. The tumors showed a highly heterogeneous composition: areas of dense presence of a single PCR product was verified by melting point vascularization were interspersed with white necrotic foci and analysis. After the exclusion of the outliers, means of three qPCR hemorrhagic spots (Fig. 1c). experiments were used for plot generation (IL-6, n ϭ 5 patients and Glioma growth on the CAM also was accompanied by a large

1644 ͉ www.pnas.org͞cgi͞doi͞10.1073͞pnas.0408622102 Hagedorn et al. Downloaded by guest on September 30, 2021 Fig. 2. Expression of VEGFR-2 during experimental glioma development. (a) During the second day after implantation, the tumor becomes vascularized from the base by CAM vessels positive for the transcript of VEGFR-2. (b)ByD4, the tumor is almost completely invaded by VEGFR-2-positive capillaries. (c)At D7, overall VEGFR-2 reactivity decreases. (d–f). Higher magniﬁcation shows endothelial cells positive for VEGFR-2 (arrows). (Scale bars: a–c, 0.5 mm; and d–f, 0.1 mm.)

peritumoral edema. As early as D2, the CAM adjacent to the tumor implantation site was visibly swollen, about six times more than controls (6.5 Ϯ 0.8 mm2 vs. 1.1 Ϯ 0.2 mm2; n ϭ 6 CAMs per group; Ͻ ϭ ϭ Fig. 3. Immunohistological and confocal microscopy analysis of experimen- P 0.001; t 6.813; df 10) (Fig. 1e). tal glioma. (a) Frozen sections of D7 tumors were stained with antivimentin We next investigated whether blood vessels inside the tumor were antibody (tumor cells, T), antidesmin antibody (pericytes, P), and fluorescein- formed through angiogenesis, a requirement for human tumor coupled SNA-1 lectin (endothelial cells, E). Note the high tumor cell density, progression (13). Expression of VEGFR-2 is up-regulated in im- the irregular and dilated vessels (arrows), and pericyte coating of most cap- mature, new blood vessels during glioma development (14). From illaries. (b) Tumor capillaries form glomeruloid-like microvascular prolifera- D2 on, capillaries invading the growing glioma were positive for tions, covered by pericytes. (c) The presence of intercapillary pillars (arrows) in transcripts of VEGFR-2, as revealed by in situ hybridization (n ϭ tumor capillaries shows intussusceptive angiogenesis. (d) Cooption of larger, ͞ dilated CAM blood vessels at the base of the tumor. (e) Flexion and orientation 4 4 tumors) (Fig. 2 a–c). At higher magnification, VEGFR-2 of a coopted vessel toward the tumor. (f) The presence of TN-C in the mRNA could specifically be detected in capillary endothelial cells extracellular matrix of glioma identified by species-specific mAbs. (g) Low- (Fig. 2 d–f). Vessels in the surrounding CAM were negative for power magnification illustrates the invasive behavior of the tumor (arrows VEGFR-2. By D4, the whole tumor was invaded by angiogenic, mark the borders of the CAM). (h) Individual glioma cells invade the CAM VEGFR-2-positive capillaries (n ϭ 2͞2 tumors) (Fig. 2b). At adjacent to the implantation site and migrate along blood vessels. (i) Exper- D7, VEGFR-2 signal was still present in capillaries throughout the imental glioma do not exhibit apoptosis as revealed by physiological distri- tumor (n ϭ 4͞4 tumors). bution of cytochrome c around cell nuclei (arrows). (Scale bars: a, d, and e, 100 ␮m; b and f,50␮m; c,20␮m; and i,10␮m.) Experimental glioma at D7 were further analyzed by immunohistochemistry and confocal microscopy. SNA-1 staining confirmed the presence of dilated tumor capillaries at the periphery of the throughout development had intact nuclear morphology as evi- tumor (Fig. 3a). Most angiogenic tumor capillaries were covered by denced by DAPI staining (data not shown) and showed a punctuate pericytes. Coverage was heterogeneous and some desmin-positive perinuclear distribution of cytochrome c (Fig. 3i), consistent with ‘‘hot spots’’ were observed, which correspond to unregulated cap- the absence of apoptosis. illary growth areas resembling glomeruloid microvascular prolifer- ations, which are typical of GBM (Fig. 3b). Two other mechanisms Treatment of Experimental Glioma with Small-Molecule Receptor contributing to tumor vascularization were further evidenced in the Tyrosine Kinase Inhibitors. We next tested whether selective blocking course of experimental glioma progression: intussusceptive micro- of growth factor receptors important for glioma progression, like vascular growth (15) (Fig. 3c) and vessel cooption (16). Invading VEGFR-2 (20) and platelet-derived growth factor receptors (21), glioma cells grow around normal CAM vessels, which subsequently could result in inhibition of experimental glioma. become part of the tumor vasculature (Fig. 3 d and e). Local treatment of size-matched glioma from D2 to D6 with ␮ The extracellular matrix in human glioma is rich in the promi- single daily doses (50 g per egg) of Gleevec, an antagonist of MEDICAL SCIENCES gratory glycoprotein TN-C (17), and TN-C expression has been platelet-derived growth factor receptor function (22) or associated with a poor clinical outcome (18). We detected abundant PTK787͞ZK 222845 (23) markedly inhibited tumor growth as early immunoreactivity for TN-C of human and chick origin in the as 2 days after the first application. Neither of the two molecules had extracellular matrix of the tumor by using species-specific mAbs visible effects on normal CAMs compared with solvent-treated (Fig. 3f). controls, which displayed a very light fibrosis, but normal Glioma cell invasion into the CAM was clearly visible on vascularization. immuno-stained sections. Blood vessels served as ‘‘guiding’’ struc- Regular vascularization, which consistently occurred in tures for migration: numerous, elongated tumor cells were tightly solvent-treated controls (n ϭ 23), was never observed in wrapped around these vessels (Fig. 3 g and h). Gleevec-treated tumors (n ϭ 22), which appeared white because Escape from apoptosis is typical of malignant tumors (13). We of a decrease in intratumoral blood flow and increased necrosis examined whether this important characteristic was preserved (Fig. 4a). Antivimentin reactivity was clearly decreased, con- during experimental glioma growth. Permeabilization of mitochon- firming reduced tumor cell density (Fig. 4 b and c). Tumor vessel drial membranes during apoptosis leads to the release of cyto- structure was markedly altered: dilated vessels, typical of control chrome c from the intermembrane space (19). Cytochrome c tumors, were not detected, and capillary lumina were irregular. release together with nuclear fragmentation are reliable reporters Pericyte coating of angiogenic capillaries was reduced, as evi- of cells undergoing apoptosis. Almost all glioma cells inside tumors denced by a decrease in desmin immunoreactivity (Fig. 4c).

Hagedorn et al. PNAS ͉ February 1, 2005 ͉ vol. 102 ͉ no. 5 ͉ 1645 Downloaded by guest on September 30, 2021 Fig. 4. Topical treatment of experimental glioma with receptor tyrosine kinase inhibitors Gleevec and PTK787͞ZK 222584. (a) Biomicros- copy follow-up of control and treated tumors. Control tumors show progressive growth and vascularization, whereas Gleevec- and PTK787͞ ZK 222584-treated tumors appear white and show signs of tissue damage. (b–d) Confocal microscopy analysis of triple-stained tumors at D7. Tumor cells (T) are blue, pericytes (P) are red, and the endothelium (E) is green. (b) Control tumors show densely packed tumor cells within a well established pericyte-covered vascular network. (c) Gleevec-treated tumors show a strong decrease in tumor cell, pericyte, and vessel density. The increase in green fluorescence is caused by nonspecific lectin binding to necrotic cells. (d)In PTK787͞ZK 222584-treated tumors, only a few pericyte-covered capillaries are found after treatment, and almost all tumor cells have dis- appeared. As in Gleevec-treated tumors, nonspecific green fluorescence increases. (e) Quantifi- cation of receptor tyrosine kinase inhibitor effects on glioma development. Bars indicate means of relative cell density and vessels counts. (Magnification: a, ϫ10; scale bar: b–d, 100 ␮m.)

Statistical analysis of Gleevec effects confirmed a significant molog to CYR61), and very low density lipoprotein͞vitellogenin decrease in relative tumor cell density (27.47 Ϯ 5.28% vs. receptor (homolog to very low density lipoprotein receptor) genes controls: 62.54 Ϯ 6.64%; mean Ϯ SEM), pericyte density (8.7 Ϯ shared sufficient similarity (Ͼ75%) for hybridization (24), so that 1.87% vs. controls: 23.3 Ϯ 4.84%), and significant reduction of the contribution of chick mRNA to the microarray signal could not angiogenic blood vessels (33.7 vessels Ϯ 9.8 vs. controls: 88.3 be excluded (Table 2). Four genes were studied in more detail. IL-6 vessels Ϯ 3.6; P Ͻ 0.05 for all comparisons, Fig. 4e). and CYR61 have important functions during human GBM devel- Treatment of experimental glioma with PTK787͞ZK 222584 also opment (25–27). Two other genes, LUM and FBXO6, are promis- abrogated progressive vascularization, as evidenced by in vivo ing, yet uncharacterized candidates that might have distinct roles in ϭ biomicroscopy (n 8 tumors) (Fig. 4d). Confocal microscopy of tumor progression (28, 29). triple-stained tumor sections revealed almost complete absence of Differential expression levels of these four genes were deter- vimentin-positive tumor cells and strong reduction of blood vessels. mined by real-time qPCR in the same reverse-transcribed RNAs Only capillaries covered by pericytes in the tumor center were used for microarray hybridization, as well as in a second series of ͞ visible after PTK787 ZK 222584 treatment (Fig. 4a). samples from an independent experimental glioma assay. Tumor Statistical evaluation of PTK787͞ZK 222584 treatment con- Ϯ cell-specific expression regulation between D2 and D4 was confirmed a significant decrease in relative tumor cell density (11.16 firmed by using human-specific primers for all four genes in the two 9.44% vs. controls: 62.54 Ϯ 6.64%; P Ͻ 0.05) and blood vessel assays (LUM: 13.9͞8-fold; CYR61:4.5͞2.3-fold; IL-6:2.3͞5.3-fold, number (10.2 vessels Ϯ 5.34 vs. controls: 88.3 vessels Ϯ 3.6; P Ͻ and FBXO6: Ϫ1.5͞Ϫ1.7-fold). 0.05), but not pericyte density (13.86 Ϯ 5.55% vs. controls: 23.3 Ϯ 4.84%; not significant) (Fig. 4e). Gene Regulation in Experimental Glioma Predicts Expression Profiles in GBM Patients. World Health Organization grade II AS tumors Gene Expression Profiling During Experimental Glioma Development. Fifty-mer oligonucleotide microarrays encompassing one-fourth of have the potential to progress to GBM (secondary GBM). ASII the human genome were used to determine gene expression tumors are poorly vascularized compared with highly angiogenic changes during the transition from early-stage tumors at D2 to fully GBMs. Progression from ASII tumors to GBMs therefore might vascularized tumors at D4. Twenty-one genes were differentially involve molecular mechanisms that are replicated in the course of regulated (Ͼ1.5-fold) between D2 and D4. Noteworthy, 6 of 21 experimental glioma. We compared expression levels of LUM, identified genes have established functions in human GBM pro- CYR61, IL-6, and FBXO6 between human GBM samples and ASII. gression (IL-6, IL-8, CYR61, MYC, MYBL2, NT5E, all up-regulated, Expression levels of individual GBM samples were evaluated by Table 1 and Table 2, which is published as supporting information qPCR for each of the four genes and normalized against the on the PNAS web site) or are produced by malignant glial cells corresponding means calculated from the ASII pool (Fig. 5). All (CEBPG). To verify whether the expression changes could specif- tumors tested for LUM showed up-regulation (12͞12, median: 7.3), ically be assigned to the human tumor cells, oligonucleotide se- with a distribution ranging from 1.3- to 36.7-fold. Most GBMs quences corresponding to the 21 genes on the array were compared showed CYR61 overexpression (8͞12, median: 3.54), up to 42-fold. with the chick genome database by using BLAST. For 18 genes, no IL-6 was also up-regulated in all GBM samples (5͞5, median: 11.55) significant homology was found, so that hybridization of avian and showed the highest relative transcription level, 143-fold. Down- transcripts could be excluded. Only avian c-MYC, CEF-10 (ho- regulation of FBXO6, as predicted by the model, was confirmed in

1646 ͉ www.pnas.org͞cgi͞doi͞10.1073͞pnas.0408622102 Hagedorn et al. Downloaded by guest on September 30, 2021 Table 1. Gene expression proﬁling of the angiogenic switch in experimental glioma by using MWG human 30K A oligonucleotide microarrays GenBank Expression levels, accession no. Gene mean Ϯ SEM

Angiogenesis NM࿝000600 IL-6 3.14 ؎ 0.37 NM࿝000584 IL-8 2.28 Ϯ 0.25 NM࿝001554 CYR61 1.86 ؎ 0.15 Cell growth control NM࿝002467 MYC͞v-myc myelocytomatosis viral oncogene homolog (avian) 2.37 Ϯ 0.06 NM࿝016447 MPP6͞maguk protein p55t; protein associated with lins 2 1.83 Ϯ 0.18 NM࿝002466 MYBL2͞v-myb myeloblastosis viral oncogene homolog (avian)-like 2 1.69 Ϯ 0.04 NM࿝002345 LUM 1.64 ؎ 0.06 NM࿝002526 NT5E͞Ecto-5Ј nucleotidase (CD73) 1.59 Ϯ 0.02 NM࿝001806 CEBPG͞CCAAT enhancer binding protein gamma (C͞EBP) 1.55 Ϯ 0.02 NM࿝021145 DMTF1͞cyclin D binding myb-like transcription factor 1 Ϫ2.23 Ϯ 0.30 NM࿝014553 TFCP2L1͞transcription factor CP2-like 1 (LBP-9; CRTR-1) Ϫ1.95 Ϯ 0.13 NM࿝018438 FBXO6 ؊1.86 ؎ 0.20 NM࿝002891 RASGRF1͞ras protein-speciﬁc guanine nucleotide-releasing factor 1 Ϫ1.82 Ϯ 0.16 Neuronal differentiation NM࿝003383 VLDLR͞very low density lipoprotein receptor 1.67 Ϯ 0.10 Immune system L17330 6H9A͞pre-T͞NK cell associated protein Ϫ2.02 Ϯ 0.18 NM࿝001276 CHI3L1͞chitinase 3-like 1 (cartilage glycoprotein-39) Ϫ1.92 Ϯ 0.25 NM࿝006187 OAS3͞2Ј-5Јoligoadenylate synthetase 3, 100 kDa Ϫ1.76 Ϯ 0.05 Others NM࿝020638 FGF23͞ﬁbroblast growth factor 23 Ϫ2.27 Ϯ 0.22 NM࿝000789 ACE͞angiotensin I converting enzyme Ϫ1.84 Ϯ 0.24 NM࿝018671 SMAP1͞hypothetical protein; iro039700 Ϫ1.75 Ϯ 0.12 NM࿝005588 MEP1A͞meprin a, alpha (PABA peptide hydrolase) Ϫ1.61 Ϯ 0.09

Twenty-one of 9,850 genes were found to be consistently regulated (Ͼ1.5-fold in three hybridization experiments) during the 48-h time lapse. Note that about one-third of regulated genes are known to be linked to glioma progression and angiogenesis, including IL-6, IL-8, and CYR61. Four selected genes that have been studied in more detail are marked in bold.

seven of nine GBMs (median: 0.7), down to 0.18 compared with In this article, we provide evidence that defined tumor growth ASII tumors. with key features of human glioblastoma at cellular and molecular levels occurs in a highly reproducible manner after human GBM Discussion cell grafting on the CAM (experimental glioma). The CAM itself has been the subject of numerous studies address- Preclinical in vivo testing of anticancer drugs is performed mainly ing activity of proangiogenic or antiangiogenic factors, but even in adult rodents, which raises major ethical concerns. Data provided though the first tumor transplantation studies were performed here strongly suggest that experimental glioma grown on the CAM Ϸ100 years ago (30) they are generally rare, which is surprising exhibit sufficient similarity with fundamental aspects of the human given the ease of access and the natural immunodeficient environ- disease. Topical treatment of experimental glioma with PTK787͞ZK 222584 for 5 days potently inhibited progressive ment of the developing bird (for review see ref. 31). vascularization, associated with a strong decrease in glioma cell density. Most likely, these effects can be attributed to functional blocking of the avian VEGFR-2, which is strongly up-regulated in the course of experimental glioma progression. Gleevec-treated tumors also exhibited consistent destruction of all cellular components of the tumor. Proliferation of U87 glioma cells depends on constitutively active platelet-derived growth factor receptors (PDGFR) signaling, which is inhibited by Gleevec (22), MEDICAL SCIENCES and PDGFR-mediated recruitment of pericytes to angiogenic capillaries is crucial for vessel stability (32). These results demonstrate that the model allows successful testing of new anticancer drugs, designed to interfere with distinct molecular pathways important for tumor progression. Experimental glioma display changes in gene expression similar to human GBM progression. IL-6 is produced by GBM cells in vitro and in vivo (33), and elevated IL-6 expression levels distinguish GBMs from low-grade AS (34). Recently, the prerequisite role of IL-6 for gliomagenesis has been demonstrated by crossbreeding GFAP-src mice, which develop GBM-like lesions, with IL-6- Fig. 5. Expression levels of four selected genes identiﬁed in the microarray deficient mice, resulting in mice without GBMs (35). The function screen were tested in patient GBM samples. Log-transformed expression levels of IL-6 in tumor progression can be assigned to its proangiogenic (x-fold over ASII pool) are shown for each patient GBM. LUM, IL-6, CYR61, and properties (36). Up-regulation of IL-6 during the angiogenic switch FBXO6 follow the same expression tendency in GBM patients compared with in experimental glioma therefore fits well with current concepts on AS experimental glioma on D4 compared with D2. Bar lines indicate median. the role of IL-6 in gliomagenesis.

Hagedorn et al. PNAS ͉ February 1, 2005 ͉ vol. 102 ͉ no. 5 ͉ 1647 Downloaded by guest on September 30, 2021 CYR61, which has a well established role in breast cancer experimental glioma growth by microarray and qPCR. More im- progression, invasion, and angiogenesis (37), was up-regulated portantly, 9͞12 patients with GBM had decreased levels of FBXO6 during experimental glioma development. CYR61 is overexpressed protein as compared with ASII, leading to the hypothesis that in 66% of patient GBMs compared with normal brain, and trans- yet-unidentified target protein(s) of FBXO6 accumulate in GBM fection of CYR61 into glioma cells resulted in faster tumor growth, and might contribute to tumor progression. The potential implica- associated with increased angiogenesis (27). The fact that up- tion of FBXO6 in the pathogenesis of CNS tumors is underscored regulation of a second, emerging GBM-progression factor has been by the recent findings that neuroblastoma with loss of heterozy- detected in our short-term model indicates that gene regulation of gosity on chromosome 1p36 are associated with down-regulation of processes typical for GBM progression, such as angiogenesis, are FBXO6 (29), to a similar level as D4 compared with D2 experi- mimicked. mental glioma. It would be interesting to see whether recently Interestingly, we confirmed up-regulation of this gene in a established experimental neuroblastoma grown on the CAM re- variable manner in one-third of patient GBMs (8͞12), a result produce this kind of gene regulation (44). similar to that of Xie et al. (27). The lack of CYR61 up-regulation We have simulated expression of significant cancer-related genes or even its down-regulation in some GBM patients may identify a in a defined 48-h time interval in vivo. The critical time frame subset of GBM with better prognosis (26). chosen for gene expression studies most likely favored identification LUM and FBXO6, differentially regulated between D2 and D4, of glioma progression genes. The angiogenic switch, a hallmark of have yet unknown biological roles in glioma progression. LUM has tumor progression, was clearly identified in experimental glioma. been associated with tumor progression and invasion (28, 38), but The fast and reproducible development of experimental glioma recent insights also suggest a role as a negative growth regulator for with key features of the human disease makes the model a powerful melanoma cells (39). Invasion of glioma cells into the CAM is very tool for preclinical in vivo studies, in an ethically favorable context. active during the first days after implantation; LUM thus may Experimental glioma is a hybrid tissue, in which only tumor participate in the regulation of the invasive process. Interestingly, all cells are of human origin. It therefore should allow selective patient GBMs also showed overexpression of LUM when compared studies of gene expression during tumor progression by the with ASII, suggesting a possible role of LUM in human GBM simultaneous use of species-specific probes, like human and pathogenesis. This finding further provides evidence that the time chick microarrays and qPCR primers. This model opens up the frame chosen for microarray analysis of experimental glioma mim- possibility to quickly screen for gene regulation occurring in the ics a significant step in human GBM progression and validates the course of new anticancer treatments (e.g., chemoresistance) in a model as a rapid screening tool for identification of genes associated highly manageable environment. with human GBM growth. FBXO6 encodes a member of the F-box protein family. F-box We thank Prof. Mathias Chiquet (University of Bern, Bern, Switzerland) proteins are variable components of the Skp1͞Cullin͞F-box-type for providing TN-C antibodies, Prof. Joerg Wilting (University of E3 ubiquitin ligase complex that function as receptors for phos- Goettingen, Goettingen, Germany) for helpful suggestions during this phorylated proteins (40). Binding of F-box proteins to their target project, Dr. Francesca De Giorgi-Ichas (Institut Europeen de Chimie et triggers a cascade, leading to the controlled degradation of the Biologie, Pessac, France) for help with the apoptosis analysis, Dr. target protein by the proteasome. Interference with this process Franc¸oiseDellu-Hagedorn (Centre National de la Recherche Scienti- results in increased levels of critical proteins like cyclin E, Notch, or fique, Bordeaux, France) for statistical analysis, Prof. Hugues Loiseau (University Bordeaux 2, Bordeaux, France) for providing patient sam- c-Myc and suggests that F-box proteins can function as tumor ples, and Drs. Jeannette Wood and Elisabeth Buchdungen (Novartis) for suppressors (41–43). Decreased expression of a given F-box protein kindly providing PTK787͞ZK 222584 and Imatinib (Gleevec). This work thus may favor tumor progression. Additionally, dosage of F-box was supported by grants from the European Union (FP6, LSHC-CT- proteins may be critical for blood vessel formation (42). We 2003-5032, STROMA Consortium) and the Ligue Nationale Contre le observed consistent decrease in FBXO6 mRNA levels during Cancer, Equipe Labellise´e (to A.B.).

1. Maher, E. A., Furnari, F. B., Bachoo, R. M., Rowitch, D. H., Louis, D. N., Cavenee, W. K. 24. Kane, M. D., Jatkoe, T. A., Stumpf, C. R., Lu, J., Thomas, J. D. & Madore, S. J. (2000) & DePinho, R. A. (2001) Genes Dev. 15, 1311–1333. Nucleic Acids Res. 28, 4552–4557. 2. Dai, C. & Holland, E. C. (2001) Biochim. Biophys. Acta 1551, M19–M27. 25. Weissenberger, J., Loeffler, S., Kappeler, A., Kopf, M., Lukes, A., Afanasieva, T. A., Aguzzi, 3. Holland, E. C., Celestino, J., Dai, C., Schaefer, L., Sawaya, R. E. & Fuller, G. N. (2000) Nat. A. & Weis, J. (2004) Oncogene 23, 3308–3316. Genet. 25, 55–57. 26. Xie, D., Yin, D., Wang, H. J., Liu, G. T., Elashoff, R., Black, K. & Koeffler, H. P. (2004) 4. Reilly, K. M., Loisel, D. A., Bronson, R. T., McLaughlin, M. E. & Jacks, T. (2000) Nat. Genet. Clin. Cancer Res. 10, 2072–2081. 26, 109–113. 27. Xie, D., Yin, D., Tong, X., O’Kelly, J., Mori, A., Miller, C., Black, K., Gui, D., Said, J. W. 5. Aldhous, P., Coghlan, A. & Copley, J. (1999) New Sci. 162, 26–31. & Koeffler, H. P. (2004) Cancer Res. 64, 1987–1996. 6. Russell, W. M. S. & Burch, R. L. (1992) The Principles of Humane Experimental Technique 28. Naito, Z., Ishiwata, T., Kurban, G., Teduka, K., Kawamoto, Y., Kawahara, K. & Sugisaki, (Universities Federation for Animal Welfare, Potters Bar, U.K.). Y. (2002) Int. J. Oncol. 20, 943–948. 7. Sherman, A., Dawson, A., Mather, C., Gilhooley, H., Li, Y., Mitchell, R., Finnegan, D. & 29. Janoueix-Lerosey, I., Novikov, E., Monteiro, M., Gruel, N., Schleiermacher, G., Loriod, B., Sang, H. (1998) Nat. Biotechnol. 16, 1050–1053. Nguyen, C. & Delattre, O. (2004) Oncogene 23, 5912–5922. 8. Pekarik, V., Bourikas, D., Miglino, N., Joset, P., Preiswerk, S. & Stoeckli, E. T. (2003) Nat. 30. Murphy, J. B. & Rous, P. (1912) J. Exp. Med. 15, 119–132. Biotechnol. 21, 93–96. 31. Hagedorn, M. & Wilting, J. (2004) in Methods in Endothelial Cell Biology, ed. Augustin, H. 9. Hagedorn, M., Balke, M., Schmidt, A., Bloch, W., Kurz, H., Javerzat, S., Rousseau, B., (Springer, Berlin), pp. 247–262. Wilting, J. & Bikfalvi, A. (2004) Dev. Dyn. 230, 23–33. 32. Lindahl, P., Johansson, B. R., Leveen, P. & Betsholtz, C. (1997) Science 277, 242–245. 10. Hagedorn, M., Zilberberg, L., Wilting, J., Canron, X., Carrabba, G., Giussani, C., Pluderi, 33. Van Meir, E., Sawamura, Y., Diserens, A. C., Hamou, M. F. & de Tribolet, N. (1990) Cancer M., Bello, L. & Bikfalvi, A. (2002) Cancer Res. 62, 6884–6890. Res. 50, 6683–6688. 11. Eichmann, A., Marcelle, C., Breant, C. & Le Douarin, N. M. (1993) Mech. Dev. 42, 33–48. 34. Rolhion, C., Penault-Llorca, F., Kemeny, J. L., Lemaire, J. J., Jullien, C., Labit-Bouvier, C., 12. Schenk, S., Muser, J., Vollmer, G. & Chiquet-Ehrismann, R. (1995) Int. J. Cancer 61, 443–449. Finat-Duclos, F. & Verrelle, P. (2001) J. Neurosurg. 94, 97–101. 13. Hanahan, D. & Weinberg, R. A. (2000) Cell 100, 57–70. 35. Tchirkov, A., Rolhion, C., Bertrand, S., Dore, J. F., Dubost, J. J. & Verrelle, P. (2001) Br. J. 14. Plate, K. H., Breier, G., Millauer, B., Ullrich, A. & Risau, W. (1993) Cancer Res. 53, 5822–5827. Cancer 85, 518–522. 15. Patan, S., Munn, L. L. & Jain, R. K. (1996) Microvasc. Res. 51, 260–272. 36. Motro, B., Itin, A., Sachs, L. & Keshet, E. (1990) Proc. Natl. Acad. Sci. USA 87, 3092–3096. 16. Holash, J., Maisonpierre, P. C., Compton, D., Boland, P., Alexander, C. R., Zagzag, D., 37. Tsai, M. S., Bogart, D. F., Castaneda, J. M., Li, P. & Lupu, R. (2002) Oncogene 21, 8178–8185. Yancopoulos, G. D. & Wiegand, S. J. (1999) Science 284, 1994–1998. 38. Leygue, E., Snell, L., Dotzlaw, H., Hole, K., Hiller-Hitchcock, T., Roughley, P. J., Watson, 17. Chiquet-Ehrismann, R. (1993) Semin. Cancer Biol. 4, 301–310. P. H. & Murphy, L. C. (1998) Cancer Res. 58, 1348–1352. 18. Herold-Mende, C., Mueller, M. M., Bonsanto, M. M., Schmitt, H. P., Kunze, S. & Steiner, 39. Vuillermoz, B., Khoruzhenko, A., D’Onofrio, M. F., Ramont, L., Venteo, L., Perreau, C., H. H. (2002) Int. J. Cancer 98, 362–369. Antonicelli, F., Maquart, F. X. & Wegrowski, Y. (2004) Exp. Cell Res. 296, 294–306. 19. Kroemer, G. & Reed, J. C. (2000) Nat. Med. 6, 513–519. 40. Kipreos, E. T. & Pagano, M. (2000) Genome Biol. 1, 3002.1–3002.7. 20. Millauer, B., Shawver, L. K., Plate, K. H., Risau, W. & Ullrich, A. (1994) Nature 367, 576–579. 41. Moberg, K. H., Bell, D. W., Wahrer, D. C., Haber, D. A. & Hariharan, I. K. (2001) Nature 21. Strawn, L. M., Mann, E., Elliger, S. S., Chu, L. M., Germain, L. L., Niederfellner, G., Ullrich, 413, 311–316. A. & Shawver, L. K. (1994) J. Biol. Chem. 269, 21215–21222. 42. Tetzlaff, M. T., Yu, W., Li, M., Zhang, P., Finegold, M., Mahon, K., Harper, J. W., Schwartz, 22. Kilic, T., Alberta, J. A., Zdunek, P. R., Acar, M., Iannarelli, P., O’Reilly, T., Buchdunger, R. J. & Elledge, S. J. (2004) Proc. Natl. Acad. Sci. USA 101, 3338–3345. E., Black, P. M. & Stiles, C. D. (2000) Cancer Res. 60, 5143–5150. 43. Yada, M., Hatakeyama, S., Kamura, T., Nishiyama, M., Tsunematsu, R., Imaki, H., Ishida, 23. Wood, J. M., Bold, G., Buchdunger, E., Cozens, R., Ferrari, S., Frei, J., Hofmann, F., N., Okumura, F., Nakayama, K. & Nakayama, K. I. (2004) EMBO J. 23, 2116–2125. Mestan, J., Mett, H., O’Reilly, T., et al. (2000) Cancer Res. 60, 2178–2189. 44. Hecht, M., Papoutsi, M., Tran, H. D., Wilting, J. & Schweigerer, L. (2004) Cancer Res. 64, 6109–6118.

1648 ͉ www.pnas.org͞cgi͞doi͞10.1073͞pnas.0408622102 Hagedorn et al. Downloaded by guest on September 30, 2021