Identification of Foxr2 As an Oncogene in Medulloblastoma

Published OnlineFirst March 5, 2014; DOI: 10.1158/0008-5472.CAN-13-1523

Cancer Tumor and Stem Cell Biology Research

Identiﬁcation of FoxR2 as an Oncogene in Medulloblastoma

Hideto Koso1,2, Asano Tsuhako1, Eli Lyons1, Jerrold M. Ward2, Alistair G. Rust3, David J. Adams3, Nancy A. Jenkins2,4, Neal G. Copeland2,4, and Sumiko Watanabe1

Abstract Medulloblastoma is the most common pediatric brain tumor, and in 25% of cases, it is driven by aberrant activation of the Sonic Hedgehog (SHH) pathway in granule neuron precursor (GNP) cells. In this study, we identified novel medulloblastoma driver genes through a transposon mutagenesis screen in the developing brain of wild-type and Trp53 mutant mice. Twenty-six candidates were identified along with established driver genes such as Gli1 and Crebbp. The transcription factor FoxR2, the most frequent gene identified in the screen, is overexpressed in a small subset of human medulloblastoma of the SHH subtype. Tgif2 and Alx4, 2 new putative oncogenes identified in the screen, are strongly expressed in the SHH subtype of human medulloblastoma. Mutations in these two genes were mutually exclusive with mutations in Gli1 and tended to cooccur, consistent with involvement in the SHH pathway. Notably, Foxr2, Tgif2, and Alx4 activated Gli-binding sites in cooperation with Gli1, strengthening evidence that they function in SHH signaling. In support of an oncogenic function, Foxr2 overexpression transformed NIH3T3 cells and promoted proliferation of GNPs, the latter of which was also observed for Tgif2 and Alx4. These findings offer forward genetic and functional evidence associating Foxr2, Tgif2, and Alx4 with SHH subtype medulloblastoma. Cancer Res; 74(8); 1–11. 2014 AACR.

Introduction of the dorsal brainstem (2). Although mutations in the SHH and Medulloblastoma is the most common malignant brain WNT pathway seem mutually exclusive, these alterations affect tumor of childhood and tends to metastasize throughout the only 30% to 40% of medulloblastomas (3), suggesting that other central nervous system. Despite overall improvements in sur- pathways are also operative in medulloblastoma formation. To vival with multimodal treatment, a substantial proportion of better understand the pathogenesis of medulloblastomas, a patients are still incurable. Moreover, survivors often suffer comprehensive understanding of genes and pathways that considerable treatment-related morbidities, such as neurocog- regulate tumor formation in the cerebellum is essential. nitive deficits related to radiation therapy. To develop better Genome-wide insertional mutagenesis is an unbiased and fi therapies, insights into the molecular mechanisms leading to high-throughput method to pro le the landscape of driver genes the development of medulloblastomas are essential. Recent in a mouse model system (4). The Sleeping Beauty (SB) transposon molecular studies have shown that medulloblastomas are system has expanded the applicability of insertional mutagenesis composed of at least four distinct subtypes: Sonic Hedgehog for the study of various types of solid tumors (5). To identify genes (SHH), WNT, Group 3, and Group 4 (1). Tumors characterized and pathways that can transform embryonic neural stem and by aberrant activation of the SHH pathway (SHH subtype) progenitor cells to brain tumor-initiating cells, we applied SB originate from cerebellar granule neuron precursor (GNP), transposon mutagenesis to the developing mouse brain. A whereas tumors with activating mutations in the WNT effector previous SB mutagenesis study screened medulloblastoma can- CTNNB1 (WNT subtype) arise outside the cerebellum from cells didate genes by analyzing recurrent insertions in medulloblastomas that developed on Ptch1 heterozygous or Trp53 (þ/ and /) mutant genetic backgrounds (6) and identified genes that Authors' Affiliations: 1Division of Molecular and Developmental Biology, promote metastatic dissemination of medulloblastomas (7). In The Institute of Medical Science, The University of Tokyo, Tokyo, Japan; contrast, we used wild-type or Trp53 mutant (R172H/þ)genetic 2 Division of Genetics and Genomics, Institute of Molecular and Cell Biol- backgrounds that do not develop medulloblastomas spontane- ogy, Agency for Science, Technology and Research, Singapore, Singa- pore; 3Experimental Cancer Genetics, Wellcome Trust Sanger Institute, ously. Transposition of the mutagenic SB transposons induced Hinxton, Cambridge, United Kingdom; and 4Cancer Research Program, medulloblastomas in the cerebellum even on the wild-type The Methodist Hospital Research Institute, Houston, Texas background. Here, we describe the identification and functional Note: Supplementary data for this article are available at Cancer Research analyses of candidate genes that may regulate the initiation Online (http://cancerres.aacrjournals.org/). and/or progression of medulloblastoma. Corresponding Authors: Sumiko Watanabe, Institute of Medical Science, University of Tokyo, 4-6-1 Shirokanedai, Minato-ku, Tokyo 108-8639, Japan. Phone: 81-3-5449-5663; Fax: 81-3-5449-5474; E-mail: Materials and Methods [email protected]; and Neal G. Copeland, [email protected] Mouse strains doi: 10.1158/0008-5472.CAN-13-1523 Mouse strains used in this study are described previously (8). 2014 American Association for Cancer Research. All manipulations were performed with institutional animal

www.aacrjournals.org OF1

Koso et al.

care and use committee approval (Institute of Molecular and (Promega) and sequenced. Real-time PCR analysis was car- Cell Biology). ried out on Light Cycler 1.5 (Roche) using primers in Supplementary Table S1. Cell culture Mouse embryonic fibroblasts (MEF) were generated from Results ICR mouse embryos at embryonic day 13.5. Primary cerebellar Sleeping Beauty mutagenesis promotes medulloblastoma cultures were prepared from the cerebellum of ICR mice at formation in the cerebellum postnatal day 5 as described (9). Briefly, the cerebellum was To identify novel driver genes for medulloblastoma, we treated with 0.25% trypsin with 0.1% DNaseI (Invitrogen), and mobilized SB in the brains of wild-type and Trp53R172H mutant then dissociated by trituration. The dissociated cells were mice. Trp53R172H is an inducible lox-stop-lox allele with both cultured in DMEM/F12 medium supplemented with 10% FBS dominant-negative and gain-of-function properties (15). Brief- and penicillin/streptomycin. MEFs, cerebellar cells, and ly, Nestin-cre (Nes-Cre) transgenic mice (16) were crossed with NIH3T3 cells (American Type Culture Collection, ATCC) were Trp53R172H mutant mice (15) to generate compound hetero- infected with retrovirus and selected for puromycin resistance. zygous Nes-cre/þ; Trp53R172H/þ mice. These mice were then For cotransduction experiments, doubly infected cells were crossed to mice homozygous for a transposon concatemer selected with puromycin and G418. Foci formation assay and carrying up to 350 copies of the mutagenic transposon T2/ cell-proliferation assay in low serum media were performed as Onc2 (17) and an inducible LSL-SB transposase (SBase) tar- described (10). Cerebellar cells were plated to 6-well plates for geted to the Rosa26 locus (18) to generate triple (Nes-cre/þ;T2/ colony formation assay (more than 50 colonies were counted Onc2/þ; SBase/þ) and quadruple (Trp53R172H/þ; Nes-cre/þ; from three samples), or to 8-well chamber slide for immunos- T2/Onc2/þ; SBase/þ) mice. Trp53R172H was used as a sensi- taining (more than 100 Math1-positive cells were counted in tizing mutation because mutant Trp53 is one of the key drivers three samples). of medulloblastoma (19), whereas Nes-cre was used to activate Trp53R172H and SB transposition in neural stem and progenitor Immunohistochemical analysis cells. Triple transgenic mice showed enlargement of the head Immunohistochemistry was performed as described (8). at approximately 2 to 4 months of age and a gait disorder, Each tumor was obtained from a different mouse. Anti-Math1 whereas necropsy of these mice revealed tumors within the antibody is a generous gift from Dr. J. Johnson (University of cerebellum (Fig. 1A–C). In contrast, control animals (T2/Onc2/ Texas Southwestern Medical Center, Dallas TX; ref. 11). The þ; SBase/þ) remained tumor free (Fig. 1A, black line). Hema- detection of senescence-associated b-galactosidase activity toxylin and eosin (H&E) staining of tumors showed classical was performed as described (12). features of medulloblastoma, including small round tumors composed of sheets of undifferentiated cells with minimal Production of retrovirus stock cytoplasm (Fig. 1D and E). Tumors were negative for glioma Retroviral expression vectors (pMXs-IRES-Puro and pMXs- markers such as GFAP and Nestin (Fig. 1F and G), providing IRES-Neo) were obtained from Cell Biolabs, Inc. A plasmid further evidence they are medulloblastomas. Trp53R172H/þ; coding for full-length Foxr2 was obtained from OriGene. A Nes-cre/þ mice also succumbed to tumors but the tumors truncated Gli1 was excised from full-length Gli1 (OriGene) with were of skin cell origin (15). Trp53R172H mutant mice with active KpnI (located in exon 5) and XhoI to remove the N-terminal 125 SB transposition also developed medulloblastomas in the amino acids required for interaction with Sufu (13). A trun- cerebellum but at a higher frequency than in triple transgenic cated form of Alx4 was PCR-amplified using primers in Sup- mice lacking Trp53R172H (Fig. 1A, blue line; Table 1, 59% vs. 40%; plementary Table S1. A cDNA insert for Tgif2 was excised from P ¼ 0.026, Fisher exact test). PCR analysis of medulloblastoma pMX-Tgif2-IRES-EGFP (14). Retroviral production was per- genomic DNA showed that the wild-type Trp53 allele was still formed as described (8). present in tumors (Fig. 1H). This was not unexpected because Trp53R172H is a dominant negative Trp53 allele that can inac- Transfection assay tivate wild-type Trp53 (15). Together, these results indicate HEK-293T cells (ATCC) were plated into 24-well plates and that mobilized transposons can promote the formation of transfection was carried out according to the manufacturer's medulloblastoma in the cerebellum. protocol (GeneJuice; Millipore). Cells were harvested 2 days after transfection. Luciferase assay were performed according Sequence analysis of SB insertion sites in to the manufacturer's protocol (Promega). Luciferase activities medulloblastomas were normalized by total protein amounts. To identify driver genes for medulloblastoma, we PCR- amplified and sequenced the transposon insertion sites from Identification of common insertion sites and the analysis 17 wild-type and 27 Trp53R172H mutant tumors (Table 1). In of fusion transcripts total, 7,806 and 13,057 nonduplicated SB insertions were Common insertion sites (CIS) were identified as described identified in wild-type and Trp53R172H mutant tumors, respec- (8). To analyze fusion transcripts, cDNA synthesis was tively (Table 1). Insertions were then analyzed using the GKC performed using RevertraAce (Toyobo) and subsequent PCR method (20) to identify common insertion sites (CIS). CISs are was performed using the primers listed in Supplementary regions in cancer genomes that harbor a higher number of Table S1. PCR products were TA-cloned into pGEM-Teasy transposon insertions than predicted by random chance and

OF2 Cancer Res; 74(8) April 15, 2014 Cancer Research

Medulloblastoma Cancer Gene Screen

Figure 1. Transposon mutagenesis induces medulloblastoma. A, Kaplan–Meier survival curves show that mice with mobilized transposons display shortened survival compared with mice without mobilized transposons. B, tumors were found in the cerebellum (arrows). C–E, H&E staining of tumors show classic histologic features of medulloblastomas. Arrows, multinucleated giant cells. Bars, 100 mm. F and G, tumors were negative for GFAP (F, left) and Nestin (G, left). GFAP and Nestin staining were observed in cortical astrocytes (F, right) and the ventricular zone of embryonic brain (G, right), respectively. Bars, 100 mm. H, PCR-based detection of the recombined and activated Trp53R172H mutant allele. The Trp53R172H mutant allele was detected in the cerebellum (Ce) of double transgenic animals (Nes-cre/þ;Trp53R172H/þ)but not in control (Ctrl) animals (Trp53R172H/þ). Both Trp53 WT and Trp53R172H mutant alleles were observed in 6 MBs (T1–T6).

therefore most likely to harbor candidate cancer genes. These DAVID Bioinformatics Resources (http://david.abcc.ncifcrf. CISs were then annotated to the nearest mouse gene (CIS gov/). A Gene Ontology term enrichment analysis of the genes; ref. 20). Thirteen CIS genes were identified in wild-type combined CIS gene list (26 genes in Fig. 3A) identified pro- tumors (Fig. 2A) and 18 CIS genes were identified in Trp53R172H cesses associated with transcriptional regulation as being mutant tumors (Fig. 2B). Five CIS genes, including Foxr2, Tgif2, highly significant (Fig. 2C). Ten out of 26 CIS genes (38%) are Alx4, Crebbp, and Wac, were identified in both wild-type and involved in transcription, and include transcription factors Trp53R172H mutant tumors (Fig. 3A), indicating that these (Foxr2, Tgif2, Alx4, Gli1, and Hivep3), transcription cofactors genes regulate medulloblastoma formation irrespective of the (Crebbp and Trim33), a histone acetyltransferase (Crebbp), a genetic background. histone methyltransferase (Whsc1l1), a regulator of histone To determine whether the CIS genes are enriched in sig- H2B ubiquitination (Wac), and a component of the SWI/SNF naling pathways important in cancer, we made use of the chromatin remodeling complex (Arid1b).

Table 1. Analysis of transposon insertions in medulloblastomas

Mice Tumor Number of Number of Genotype necropsya Medulloblastomab samplesc insertionsd CIS genes T2/Onc2/þ; SBase/þ (n ¼ 5) 5 0 (0%) Nes-cre/þ; T2/Onc2/þ; SBase/þ (n ¼ 43) 19 17 (40%) 17 7,806 13 Trp53R172H/þ; Nes-cre/þ (n ¼ 10) 10 0 (0%) Trp53R172H/þ; Nes-cre/þ; T2/Onc2/þ; SBase/þ (n ¼ 58) 37 34 (59%) 27 13,057 18 Total 44 20,863 26

aThe number of mice that were sent to necropsy. bThe number of mice with medulloblastoma at histologic analysis. cThe number of tumor samples used for the analysis of transposon insertions. Each tumor is from a different mouse. dInsertions in the transposon donor chromosomes were not included because of problem caused by local hopping.

www.aacrjournals.org Cancer Res; 74(8) April 15, 2014 OF3

Koso et al.

Figure 2. CIS genes identified in medulloblastomas. A, thirteen CIS genes were identified in wild-type tumors. B, eighteen CIS genes were identified in Trp53 mutant tumors. P values were calculated as described (20). Prediction on gene function indicates whether the T2/Onc2 insertion would cause transcriptional activation of a gene (drives) or would disrupt gene transcription (disrupts) on the basis of the position and orientation of the T2/Onc2 insertion relative to gene transcription. C, GO analysis of biological processes of CIS genes in medulloblastomas.

GLI1, a downstream effector of SHH signaling, is often ampli- scripts showed that the MSCV promoter was spliced to exon 2 fied in medulloblastomas of the SHH subtype (21). Gli1 was (Fig. 3J), again driving the expression of a full-length Tgif2 identified as a CIS gene in Trp53 mutant tumors (Fig. 3A). All transcript. Tgif2 is a member of the TALE homeodomain Gli1 insertions were located upstream of exon 6, in the same protein family. Alx4 belongs to the family of aristaless-like transcriptional orientation as Gli1 (Fig. 3B), suggesting that SB homeobox genes. Alx4 insertions are similarly sense-oriented might be upregulating Gli1 expression from the MSCV pro- and located upstream of the gene or within intron 1 (Fig. 3E). moter contained within the transposon, consistent with its Sequencing of the fusion transcripts showed that the MSCV known role as an oncogene. To confirm this, we PCR-amplified promoter was spliced to Alx4 exon 2 (Fig. 3K), inducing the SB–Gli1 fusion transcripts with primers that recognize the expression of a truncated protein that may initiate from an splice donor site located downstream of the MSCV promoter ATG codon located in exon 2. and the coding region of Gli1. Sequencing of the amplified In contrast, Crebbp cand Wa insertions were mostly transcripts showed that the MSCV promoter was spliced to distributed throughout the genes, and there was little ori- exon 6 of Gli1 in both tumors examined with Gli1 insertions entation bias (Fig. 3F and G). The average numbers of Crebbp (Fig. 3H), resulting in the overexpression of N-terminal trun- and Wac insertions in tumors are 1.15 and 1.18 respectively, cated transcripts (Gli1-dN) that lack the suppressor of fused suggesting a haploinsufficient role for Crebbp and Wac in (Sufu)-binding domain (13). Sufu has an inhibitory effect on Gli1 tumor suppression. Crebbp functionsasatranscriptional transcriptional activity by the sequestration of Gli1 to the coactivator, and haploinsufficiency of CREBBP is responsible cytoplasm (22). A similar truncated GLI1 is expressed in human for Rubinstein–Taybi syndrome, an autosomal congenital tumor cell lines (23). Overexpression of Gli1-dN promoted disorder characterized by developmental and pathologic focus formation in NIH3T3 cells (Fig. 5A) as well as growth phenotypes, including predisposition to tumors including in low serum (Fig. 5B). Taken together, these data indicate medulloblastoma (24, 25). Somatic heterozygous nonsense that derepression of Gli1 from Sufu-mediated suppression CREBBP mutations have been identified in medulloblasto- plays an important role in the formation of medulloblastoma. mas (26). WAC is a functional partner of histone H2B The most frequently mutated gene in medulloblastomas was ubiquitin ligase RNF20/40, and the depletion of RNF20/40 the Foxr2 transcription factor (Fig. 2A and B). Foxr2 is a or WAC abolishes H2B monoubiquitination (27). Although member of the forkhead box transcription factor family. Most significant loss of RNF20/40 is catastrophic to cells, more Foxr2 insertions are sense oriented and located upstream of subtle changes in RNF20/40 expression because of partial the protein-coding region of Foxr2 (Fig. 3C). Sequencing of the knockdown allow cells to proliferate and manifest dramatic fusion transcripts showed that the MSCV promoter was spliced genomic instability (28). These data support a haploinsuffi- to exon 2 of Foxr2 (Fig. 3I), resulting in the overexpression of a cient role for Crebbp and Wac in medulloblastomas. full-length Foxr2 transcript (Fig. 3L). We also examined the A previous transposon screen that analyzed for recurrent patterns of transposon insertions in the other four genes that insertions in medulloblastomas, which developed on a Ptch1 were identified irrespective of genetic background (Fig. 3A). heterozygous genetic background (6) did not identify Gli1, Tgif2 insertions are all sense-oriented and located upstream of Foxr2, Tgif2,orAlx4, raising the possibility that these genes the initiating codon (Fig. 3D). Sequencing of the fusion tran- have overlapping function with Ptch1 deficiency. We also

OF4 Cancer Res; 74(8) April 15, 2014 Cancer Research

Medulloblastoma Cancer Gene Screen

Figure 3. Patterns of transposon insertions in medulloblastoma candidate genes. A, thirteen and 18 CIS genes were identified in wild- type and Trp53 mutant tumors, respectively. Five CIS genes were shared between the wild-type and Trp53 mutant tumors. B–G, SB insertions are shown for wild-type and Trp53 mutant tumors for the CIS genes Gli1 (B), Foxr2 (C), Tgif2 (D), Alx4 (E), Crebbp (F), and Wac (G). Transposon insertions are located in the sense (red arrows) or antisense orientation (white arrows) relative to transcription. H–K, fusion transcripts were PCR amplified with primers designed to recognize the splice donor site located downstream of the MSCV promoter in T2/Onc2 and the coding region of Gli1 (H), Foxr2 (I), Tgif2 (J), or Alx4 (K), and sequenced. L, expression level of Foxr2 in tumors with Foxr2 insertions (n ¼ 4) and tumors without (n ¼ 4). Expression levels were normalized using b-actin as an internal control, and the average was set to a value of 1. Data are presented as box-whisker plots showing the minimum, 25th percentile, median, 75th percentile, and the maximum (P-value: Student t test, one-tailed). M, co- occurrence of insertions in Foxr2, Tgif2, Alx4, and Gli1 in 44 tumors. N, significance (P-value) of co- occurring insertions was determined using the Fisher exact test for each combination.

looked for co-occurring and mutually exclusive mutations in medulloblastoma, but in more limited subsets of tumors pairs of CIS genes. A signiﬁcant co-occurrence of mutations in (Supplementary Fig. S1C and S1D), and once again most of Tgif2 and Alx4 genes was observed (Fig. 3M and N; P ¼ 0.042 these tumors are of the SHH subgroup. We also found that after multiple testing correction), whereas Tgif2 and Alx4 most of the tumors highly expressing ALX4 also expressed insertions were mutually exclusive to Gli1 insertions (Fig. 3M). TGIF2 at high levels (Supplementary Fig. S1E), showing signiﬁcant co-occurrence of ALX4 and TGIF2 in medulloblas- Expression of GLI1, FOXR2, TGIF2, and ALX4 in human tomas (P < 0.001, Fisher exact test). Taken together, these medulloblastomas data indicate that these 4 putative oncogenes are overex- Because the patterns of transposon insertions predict pressed in a subset of human medulloblastomas that belong oncogenic functions for GLI1, FOXR2, TGIF2,andALX4 in to the SHH subgroup. medulloblastoma, we next determined whether expression of these genes was upregulated in human medulloblastoma Foxr2, Tgif2, and Alx4 enhance transcriptional activity of using a microarray dataset for 285 medulloblastomas (29) Gli1 that was accessible through the open access database R2 Because FOXR2, TGIF2, and ALX4 are expressed in a subset of (http://r2.amc.nl). Not surprisingly, GLI1 is strongly exp- SHH subgroup medulloblastomas, we next tested whether these ressed in medulloblastomas of the SHH subtype (Supplemen- genes can activate a SHH responsive element, the Gli-binding tary Fig. S1A). Interestingly, most tumors that strongly site (Gli-BS; ref. 30). Transfection of a vector expressing full- express TGIF2 also belong to the SHH subgroup (Supplemen- length Gli1 increased reporter expression in HEK-293T cells; tary Fig. S1B). ALX4 and FOXR2 are also strongly expressed in however, expression of the reporter with mutated Gli-binding

www.aacrjournals.org Cancer Res; 74(8) April 15, 2014 OF5

Koso et al.

Figure 4. Foxr2, Alx4, and Tgif2 cooperate with Gli1 and activate Gli-dependent transcription. A, expression of Gli-binding site luciferase (Gli-BS-Luc) was upregulated by transfection with Gli1 expression vector (the average luciferase activity from the Gli1 expression vector was designated as 100), whereas no activation was observed with mutated Gli-binding site luciferase (mGli-BS-Luc; ref. 30). Transcriptional activation of Gli-BS-Luc was upregulated by cotransfection with Foxr2, Alx4,or Tgif2 expression vectors. Data represent means SEM with three samples per group. The amounts of DNA (mg) used for transfection are shown. , P < 0.05 and , P < 0.01 (Student t test, one-tailed). B, the experiments shown in A were repeated three times, and data represent means SEM with 9 samples per group. , P < 0.01 (Student t test, one-tailed). C, dose-dependent cooperative effects were observed for Foxr2, Alx4, and Tgif2. Data represent means SEM with 9 samples per group. , P < 0.01 (Student t test, one-tailed). D, the expression of Ptch1 and Ptch2 in NIH3T3 cells overexpressing Foxr2, Alx4,or Tgif2. Expression was normalized using b-actin as an internal control. Data represent means SEM with three samples per group. , P < 0.05 and , P < 0.01 (Student t test, one- tailed). E, cooperation between Alx4 and Tgif2 on activation of Gli- dependent transcription. Data represent means SEM with 9 samples per group. , P < 0.01 (Student t test, one-tailed).

site (mGli-BS; ref. 30) was not changed (Fig. 4A). Transfection of along with the Gli1 expression vector increased Gli1 reporter vectors expressing Foxr2, Tgif2,orAlx4 alone did not affect expression above that seen with the Gli1 expression vector alone reporter expression; however, cotransfection of these vectors (Fig. 4A and B). This upregulation of Gli1-dependent reporter

OF6 Cancer Res; 74(8) April 15, 2014 Cancer Research

Medulloblastoma Cancer Gene Screen

Figure 5. Functional analysis of CIS genes. A–E, NIH3T3 cells were infected with retroviruses expressing Gli1-dN, Foxr2, Alx4, Tgif2, or empty vector control. These cells were used for the focus formation assay (A and D) and cell growth assay in low serum (B and E). Bars, 250 mm (C). Data represent means SEM with three samples per group. , P < 0.05 and , P < 0.01 (Student t test, one- tailed). Data are representative of two independent experiments. F–I, MEFs were infected with retroviruses expressing Foxr2 or empty vector, and serially passaged. Senescence- associated b-gal expression was examined at passage 2 (F). Foxr2- expressing MEFs showed prolonged proliferation compared with control (G). The proportion of b-gal–expressing cells in control and Foxr2-expressing MEFs (H). The expression of p16Ink4 was lower in Foxr2-expressing MEFs than in control (I). Expression was normalized using b-actin as an internal control. Data represent means SEM with three samples per group. , P < 0.05 and , P < 0.01 (Student t test, one-tailed). Data are representative of two independent experiments.

expression was observed dose dependently (Fig. 4C). As infected NIH3T3 cells with retroviruses expressing full-length expected, expression of endogenous SHH target genes in Foxr2, Tgif2, truncated Alx4, or an empty vector. Foxr2 over- NIH3T3 cells, Ptch1,andPtch2 (31) were upregulated by Foxr2, expression induced morphologic changes (Fig. 5C), and pro- Tgif2,orAlx4 (Fig. 4D). Cotransfection of Tgif2 and Alx4 duced a significantly greater number of NIH3T3 foci, whereas induced stronger luciferase reporter expression compared truncated Alx4 did not promote and Tgif2 suppressed foci with Tgif2 or Alx4 alone (Fig. 4E), providing evidence for formation (Fig. 5D). Analysis of cell growth in low-serum media functional cooperation between these genes. Taken together, also showed that Foxr2-expressing cells grew more efficiently these results demonstrate cooperative functions of these (Fig. 5E). These data indicate transforming activity of Foxr2 genes (Foxr2, Tgif2,orAlx4)andGli1 in Gli-dependent tran- in NIH3T3 cells. scriptional activation. We next analyzed the effects of Foxr2 overexpression on MEFs. Normal MEFs undergo senescence as a result of serial Functional analysis of candidate medulloblastoma passage in culture (32). As expected, control MEFs became driver genes enlarged and ceased to proliferate after several passages, and To examine the oncogenic functions of candidate genes, we these cells expressed senescence-associated b-galactosidase first measured their transforming activity in NIH3T3 cells. We (Fig. 5F and G). However, Foxr2-expressing MEFs retained their

www.aacrjournals.org Cancer Res; 74(8) April 15, 2014 OF7

Koso et al.

Figure 6. Overexpression of medulloblastoma candidate genes promotes proliferation of GNPs. A, cerebellar cells infected with retroviruses expressing Foxr2, Alx4, Tgif2, or empty vector. Arrowheads, large colonies consisting of GNPs. Bars, 250 mm. B, magniﬁed views of GNP colonies. Bars, 50 mm. C, GNP colonies contained cytoplasmic granules that were visible under differential interference contrast (DIC) microscopy (left). These granular cells expressed Math1, but were negative for S100b (right). Arrow, the nucleus of an astrocyte. Nuclei were counterstained with DAPI. Bars, 50 mm. D–F, colony size and proliferation of GNPs were compared among control, Foxr2, Alx4,andTgif2-expressing GNPs. Foxr2-expressing GNPs generated larger colonies than control (D; Mann–Whitney test, P < 0.001). The fraction of Ki67-expressing Math1-positive GNPs was increased by Foxr2 overexpression (E and F). Data represent means SEM with three samples per group. , P < 0.01 (Student t test, one-tailed). Data are representative of two independent experiments. G, cerebellar cells infected with retroviruses expressing Alx4 and Tgif2,orAlx4 and empty vector. Most of the control GNPs remained single (arrows), whereas cotransduction of Alx4 and Tgif2 induced colonies consisting of GNPs (arrowheads). Bars, 250 mm(top)and50mm (bottom). H and I, GNPs transduced with Alx4 and Tgif2 generated larger colonies than control (Mann–Whitney test, P < 0.001; H). The fraction of Ki67-expressing Math1-positive GNPs was increased by cotransduction of Alx4 and Tgif2 (I). Data represent means SEM with three samples per group. , P < 0.01 (Student t test, one-tailed). Data are representative of two independent experiments.

small cell shape and continued to proliferate (Fig. 5F and G), Cerebellar GNPs are considered to be the cell-of-origin of the and the proportion of b-galactosidase expressing cells SHH subgroup of medulloblastomas (3). We therefore next decreased (Fig. 5H). p16Ink4a, a cyclin-dependent kinase inhib- examined the effects of overexpression of Foxr2, truncated itor, is also expressed in most senescent cells (32). As expected, Alx4, and Tgif2 on cultures of primary cerebellar cells. These the expression level of p16Ink4a increased during passages of cultures consisted primarily of GNPs and astrocytes (9). At 1 MEFs; however, its expression was significantly lower in MEFs week after plating, colonies consisting of GNPs were distin- overexpressing Foxr2 (Fig. 5I). Taken together, these results guishable from surrounding large flat astrocytes because of indicate that overexpression of Foxr2 prevents MEFs from their small cell shape and cytoplasmic granules (Fig. 6A–C). entering senescence. These small granular cells expressed Math1, a specific marker

OF8 Cancer Res; 74(8) April 15, 2014 Cancer Research

Medulloblastoma Cancer Gene Screen

for GNPs (9, 11), but were negative for an astrocyte marker, Wac, providing further evidence that these genes are medul- S100b (Fig. 6C). The colony size was larger in Foxr2-expressing loblastoma disease genes. The patterns of transposon inser- GNPs than control cells (Fig. 6A, B, and D). The expression of tions predict tumor suppressor functions for these genes. In Ki67, a marker associated with cellular proliferation, was also fact, recurrent mutations in cancers have been reported for increased by Foxr2 overexpression (Fig. 6E and F). These TRIM33 (36), ARID1B (37), and MAP2K4 (38). Interestingly, findings demonstrate a growth promoting function for Foxr2 genes such as Gli1, Foxr2, Tgif2, and Alx4 were not identified in primary GNPs. Although Alx4 or Tgif2 alone did not promote in the Ptch1 and Trp53 mutant screen (7), raising the possibility proliferation (Fig. 6D and F), cotransduction of Alx4 and Tgif2 that these genes might function in SHH signaling and were induced colonies with larger sizes than Alx4 alone (Fig. 6G and therefore not identified because the SHH pathway was already H). Ki67 expression was also increased by cotransduction of activated by either Ptch1 germline or transposon-induced Alx4 and Tgif2 (Fig. 6I), providing evidence for cooperation somatic mutations. between these genes in GNPs. The most frequently mutated gene identified in our screen was Foxr2. Fox proteins can both activate and repress gene expression through the recruitment of cofactors or repressors, Discussion primarily histone deacetylases. Deregulation of Fox proteins is Transposon-based insertional mutagenesis allows for unbi- commonly associated with cancer progression (39). FOXR2 has ased, whole genome screens for cancer genes, and has been high homology to FOXR1, which can functionally replace MYC used successfully to model many different types of cancer in and drive proliferation of a neural crest cell line (40). A recent mice (5). In this study, we used transposon mutagenesis to mutagenesis screen identified Foxr2 as a proto-oncogene in model medulloblastoma. Using Nes-cre transgenic mice, we malignant peripheral nerve sheath tumors (41). This study mobilized T2/Onc2 transposons in neural stem and progenitor provides the first forward genetic evidence associating Foxr2 cells of the developing brain. As a result, mutagenic transpo- with medulloblastoma. Foxr2 insertion pattern predicts an sons induced medulloblastomas in the cerebellum. Strikingly, oncogenic role for Foxr2. Consistent with this, Foxr2 over- we observed medulloblastomas even on a wild-type genetic expression showed transforming activity in NIH3T3 cells, background. These tumors are invaluable because they have suppressed senescence in primary fibroblasts and promoted the potential to identify driver genes that can initiate tumor- the proliferation of GNPs. Interestingly, Foxr2 was strongly igenesis rather than just accelerate tumorigenesis induced by expressed in a small subset of medulloblastomas that belonged mutation of another gene, such as a mutation of Ptch1. to the SHH subgroup of medulloblastoma. Subsequent sequence analysis of the transposon integrations The analysis of co-occurring and mutually exclusive pairs of sites in these tumors identified known medulloblastoma-asso- insertions in CIS genes showed that Gli1 insertions were ciated genes Gli1 and Crebbp as well as genes that have not mutually exclusive to insertions in Tgif2 and Alx4. In addition, been associated with medulloblastoma, including Foxr2, Tgif2, we found a significant co-occurrence between insertions in Alx4, and Wac. Tgif2 and Alx4. Analysis of the expression of these genes in A previously published transposon screen identified medul- human medulloblastoma showed that TGIF2 and ALX4 are loblastoma candidate genes by analyzing for recurrent inser- strongly expressed in a subset of medulloblastomas that belong tions in medulloblastomas that developed on heterozygous to the SHH subgroup. This again raises the possibility that Tgif2 Ptch1 mutant or Trp53 mutant (þ/ and /) backgrounds and Alx4 function in SHH signaling. Consistent with this, (6). Ptch1 is a negative regulator of the SHH signaling pathway expression of Gli1 is reduced in Tgif2 conditional null mice, (33) and not surprisingly, Ptch1 heterozygous mice spontane- and deficiency of Gli3, a potent repressor of SHH signaling, ously develop medulloblastomas (34). CIS genes identified on partially rescues the Tgif2 null phenotype (42). Interestingly, the Ptch1 mutant background were shown to promote meta- the Alx4 null phenotype can also be partially rescued by loss of static dissemination of medulloblastoma (7). The previous Gli3, further suggesting that Alx4 functions in SHH signaling study (6) also identified Ptch1 in medulloblastomas induced (43). Our studies showing that TGIF2 and ALX4 genes are on Trp53 mutant backgrounds. The identification of Ptch1 on strongly expressed in the SHH subgroup of medulloblastoma the Trp53 / background is consistent with a previous report also supports this notion. describing cooperation between Trp53 loss and Ptch1 hetero- Importantly, we found that Foxr2, Tgif2, and Alx4 cooperated zygosity for medulloblastoma formation (35). In contrast, it has with Gli1 and activated Gli-dependent transcription, demon- been reported that Ptch1 heterozygosity does not accelerate strating the involvement of these genes in SHH signaling. medulloblastoma formation on Trp53 heterozygous back- Tumors with mutations in members of the SHH pathway ground (35). This may explain why Ptch1 was not identified constitute nearly 25% of medulloblastomas (26, 44) that belong in our screen on the Trp53R172H/þ heterozygous mutant to the SHH subgroup of medulloblastomas, and the aberrant background. We also used a wild-type background, which did activation of the SHH pathway is considered to be causally not develop medulloblastoma spontaneously. Therefore, some related to this subgroup. Although Alx4 and Tgif2 promoted of the CIS genes identified in our screen are likely to represent Gli-dependent transcription, these genes did not promote driver genes involved in the initiation of medulloblastoma. proliferation of GNPs by itself. In contrast, cotransduction of Comparison of our CIS genes with those identified in the Ptch1 Alx4 and Tgif2 enhanced Gli-dependent transcription and and Trp53 mutant screen showed that 5 genes were identified promoted proliferation. A previous study showed that addi- in both screens, including Arid1b, Crebbp, Trim33, Map2k4, and tional factors define an activation threshold for Gli target genes

www.aacrjournals.org Cancer Res; 74(8) April 15, 2014 OF9

Koso et al.

(45), raising a possibility that similar mechanisms may regulate tomasaswellasforcancersthatinvolveSHHsignaling. a proliferative response to the Gli transcriptional activity in Future studies will be required to determine the exact roles GNPs. The SHH signaling pathway has been implicated in a of these genes in a wide variety of human cancers. wide variety of human tumors (46). Analysis of the expression level of FOXR2, TGIF2, and ALX4 in human cancer cell lines Disclosure of Potential Conflicts of Interest using the CCLE database (47) revealed that nearly 7% of cancer No potential conflicts of interest were disclosed. cell lines showed higher expression of FOXR2 than 1.5 inter- quartile ranges above the upper quartile (Supplementary Fig. Authors' Contributions Conception and design: H. Koso, N.A. Jenkins, N.G. Copeland, S. Watanabe S1F). Interestingly, FOXR2 strongly expressing tumors include Development of methodology: N.A. Jenkins, N.G. Copeland cancers associated with SHH signaling such as multiple mye- Acquisition of data (provided animals, acquired and managed patients, loma, melanoma, and glioma (46), raising a possibility that provided facilities, etc.): H. Koso, A. Tsuhako, J.M. Ward, D.J. Adams, N.A. Jenkins, N.G. Copeland FOXR2 may also be involved in other cancers associated with Analysis and interpretation of data (e.g., statistical analysis, biostatistics, SHH signaling. ALX4 and TGIF2 were strongly expressed in 2% computational analysis): H. Koso, E. Lyons, A.G. Rust, D.J. Adams, N.A. Jenkins, and 0.1% of cancer cell lines, respectively (Supplementary Fig. N.G. Copeland, S. Watanabe Writing, review, and/or revision of the manuscript: H. Koso, E. Lyons, J.M. S1G and S1H). Although we cannot distinguish the expression Ward, N.A. Jenkins, N.G. Copeland, S. Watanabe of truncated ALX4 from that of full-length ALX4, nearly half of Administrative, technical, or material support (i.e., reporting or orga- nizing data, constructing databases): E. Lyons, N.A. Jenkins, N.G. Copeland the cell lines that strongly express ALX4 belonged to the lung Study supervision: N.A. Jenkins, N.G. Copeland non–small cell carcinoma, in which SHH signaling plays an essential role (48). Acknowledgments Interestingly, nearly 40% of genes identified in our screen The authors thank K. Rogers and S.M. Rogers for rodent necropsy. The authors were associated with transcription, consistent with a pre- also thank P. Cheok, N. Lim, and D. Chen for their help with tumor monitoring. C.C.K. Yew is acknowledged for the bioinformatics analysis of transposon insertion vious screen for medulloblastoma disease genes (49). Tran- sites. We are grateful to Dr. J. Johnson for providing the anti-Math1 antibody. scription factors play essential roles in coordinated gene expression during development. Recently, the aberration of Grant Support gene expression has been implicated in mechanisms of This work was supported by the Japan Society for the Promotion of Science, cancer formation in the pediatric setting (26, 50), raising the Biomedical Research Council, the Agency for Science, Technology and Research, Singapore, and the Cancer Prevention Research Institute of Texas the possibility that aberrations of gene expression may (CPRIT). D.J. Adams and A.G. Rust are supported by Cancer Research-UK and the perturb the normal developmental program and cause Wellcome Trust. N.G. Copeland and N.A. Jenkins are also both CPRIT Scholars in Cancer Research. deregulated proliferation of stem or progenitor cells. The The costs of publication of this article were defrayed in part by the finding of transcriptional regulators in our screen supports payment of page charges. This article must therefore be hereby marked the notion that deregulation of the developmental program advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this is a core mechanism of pathogenesis in medulloblastomas fact. fi (3). Taken together, our ndings implicate FOXR2, TGIF2, Received June 6, 2013; revised January 31, 2014; accepted February 22, 2014; and ALX4 as potential therapeutic targets for medulloblas- published OnlineFirst March 5, 2014.

References 1. Northcott PA, Korshunov A, Pfister SM, Taylor MD. The clinical 10. Park HJ, Wang Z, Costa RH, Tyner A, Lau LF, Raychaudhuri P. An N- implications of medulloblastoma subgroups. Nat Rev Neurol 2012; terminal inhibitory domain modulates activity of FoxM1 during cell 8:340–51. cycle. Oncogene 2008;27:1696–704. 2. Gibson P, Tong Y, Robinson G, Thompson MC, Currle DS, Eden C, 11. Helms AW, Johnson JE. Progenitors of dorsal commissural inter- et al. Subtypes of medulloblastoma have distinct developmental neurons are defined by MATH1 expression. Development 1998;125: origins. Nature 2010;468:1095–9. 919–28. 3. Gilbertson RJ, Ellison DW. The origins of medulloblastoma subtypes. 12. Debacq-Chainiaux F, Erusalimsky JD, Campisi J, Toussaint O. Pro- Annu Rev Pathol 2008;3:341–65. tocols to detect senescence-associated b-galactosidase (SA-bgal) 4. Kool J, Berns A. High-throughput insertional mutagenesis screens in activity, a biomarker of senescent cells in culture and in vivo. Nat mice to identify oncogenic networks. Nat Rev Cancer 2009;9:389–99. Protoc 2009;4:1798–806. 5. Copeland NG, Jenkins NA. Harnessing transposons for cancer gene 13. Dunaeva M, Michelson P, Kogerman P, Toftgard R. Characterization of discovery. Nat Rev Cancer 2010;10:696–706. the physical interaction of Gli proteins with SUFU proteins. J Biol Chem 6. Wu X, Northcott PA, Dubuc A, Dupuy AJ, Shih DJ, Witt H, et al. Clonal 2003;278:5116–22. selection drives genetic divergence of metastatic medulloblastoma. 14. Satoh S, Watanabe S. TGIF, a homeodomain transcription factor, Nature 2012;482:529–33. regulates retinal progenitor cell differentiation. Exp Eye Res 2008;87: 7. Mumert M, Dubuc A, Wu X, Northcott PA, Chin SS, Pedone CA, et al. 571–9. Functional genomics identifies drivers of medulloblastoma dissemi- 15. Olive KP, Tuveson DA, Ruhe ZC, Yin B, Willis NA, Bronson RT, et al. nation. Cancer Res 2012;72:4944–53. Mutant p53 gain of function in two mouse models of Li-Fraumeni 8. Koso H, Takeda H, Yew CC, Ward JM, Nariai N, Ueno K, et al. syndrome. Cell 2004;119:847–60. Transposon mutagenesis identifies genes that transform neural stem 16. Tronche F, Kellendonk C, Kretz O, Gass P, Anlag K, Orban PC, et al. cells into glioma-initiating cells. Proc Natl Acad Sci U S A 2012;109: Disruption of the glucocorticoid receptor gene in the nervous system E2998–3007. results in reduced anxiety. Nat Genet 1999;23:99–103. 9. Kenney AM, Rowitch DH. Sonic hedgehog promotes G(1) cyclin 17. Dupuy AJ, Akagi K, Largaespada DA, Copeland NG, Jenkins NA. expression and sustained cell cycle progression in mammalian neu- Mammalian mutagenesis using a highly mobile somatic Sleeping ronal precursors. Mol Cell Biol 2000;20:9055–67. Beauty transposon system. Nature 2005;436:221–6.

OF10 Cancer Res; 74(8) April 15, 2014 Cancer Research

Medulloblastoma Cancer Gene Screen

18. Dupuy AJ, Rogers LM, Kim J, Nannapaneni K, Starr TK, Liu P, et al. A 35. Wetmore C, Eberhart DE, Curran T. Loss of p53 but not ARF accel- modified sleeping beauty transposon system that can be used to erates medulloblastoma in mice heterozygous for patched. Cancer model a wide variety of human cancers in mice. Cancer Res 2009; Res 2001;61:513–6. 69:8150–6. 36. Hatakeyama S. TRIM proteins and cancer. Nat Rev Cancer 19. Parsons DW, Li M, Zhang X, Jones S, Leary RJ, Lin JC, et al. The 2011;11:792–804. genetic landscape of the childhood cancer medulloblastoma. Science 37. Sausen M, Leary RJ, Jones S, Wu J, Reynolds CP, Liu X, et al. 2011;331:435–9. Integrated genomic analyses identify ARID1A and ARID1B altera- 20. March HN, Rust AG, Wright NA, ten Hoeve J, de Ridder J, Eldridge M, tions in the childhood cancer neuroblastoma. Nat Genet 2013; et al. Insertional mutagenesis identifies multiple networks of cooperat- 45:12–7. ing genes driving intestinal tumorigenesis. Nat Genet 2011;43:1202–9. 38. Biankin AV, Waddell N, Kassahn KS, Gingras MC, Muthuswamy LB, 21. Zhu H, Lo HW. The human glioma-associated oncogene homolog 1 Johns AL, et al. Pancreatic cancer genomes reveal aberrations in axon (GLI1) family of transcription factors in gene regulation and diseases. guidance pathway genes. Nature 2012;491:399–405. Curr Genomics 2010;11:238–45. 39. Myatt SS, Lam EW. The emerging roles of forkhead box (Fox) proteins 22. Kogerman P, Grimm T, Kogerman L, Krause D, Unden AB, Sandstedt in cancer. Nat Rev Cancer 2007;7:847–59. B, et al. Mammalian suppressor-of-fused modulates nuclear-cyto- 40. Santo EE, Ebus ME, Koster J, Schulte JH, Lakeman A, van Sluis P, et al. plasmic shuttling of Gli-1. Nat Cell Biol 1999;1:312–9. Oncogenic activation of FOXR1 by 11q23 intrachromosomal deletion- 23. Shimokawa T, Tostar U, Lauth M, Palaniswamy R, Kasper M, Toftgard fusions in neuroblastoma. Oncogene 2011;31:1571–81. R, et al. Novel human glioma-associated oncogene 1 (GLI1) splice 41. Rahrmann EP, Watson AL, Keng VW, Choi K, Moriarity BS, Beckmann variants reveal distinct mechanisms in the terminal transduction of the DA, et al. Forward genetic screen for malignant peripheral nerve sheath hedgehog signal. J Biol Chem 2008;283:14345–54. tumor formation identifies new genes and pathways driving tumori- 24. Roelfsema JH, Peters DJ. Rubinstein-Taybi syndrome: clinical and genesis. Nat Genet 2013;45:756–66. molecular overview. Expert Rev Mol Med 2007;9:1–16. 42. Taniguchi K, Anderson AE, Sutherland AE, Wotton D. Loss of Tgif 25. Taylor MD, Mainprize TG, Rutka JT, Becker L, Bayani J, Drake JM. function causes holoprosencephaly by disrupting the SHH signaling Medulloblastoma in a child with Rubenstein-Taybi syndrome: case pathway. PLoS Genet 2012;8:e1002524. report and review of the literature. Pediatr Neurosurg 2001;35:235–8. 43. Panman L, Drenth T, Tewelscher P, Zuniga A, Zeller R. Genetic 26. Robinson G, Parker M, Kranenburg TA, Lu C, Chen X, Ding L, et al. interaction of Gli3 and Alx4 during limb development. Int J Dev Biol Novel mutations target distinct subgroups of medulloblastoma. Nature 2005;49:443–8. 2012;488:43–8. 44. Pugh TJ, Weeraratne SD, Archer TC, Pomeranz Krummel DA, 27. Zhang F, Yu X. WAC, a functional partner of RNF20/40, regulates Auclair D, Bochicchio J, et al. Medulloblastoma exome sequencing histone H2B ubiquitination and gene transcription. Mol Cell 2011;41: uncovers subtype-specific somatic mutations. Nature 2012;488: 384–97. 106–10. 28. Chernikova SB, Razorenova OV, Higgins JP, Sishc BJ, Nicolau M, 45. Wang H, Lei Q, Oosterveen T, Ericson J, Matise MP. Tcf/Lef repressors Dorth JA, et al. Deficiency in mammalian histone H2B ubiquitin ligase differentially regulate Shh-Gli target gene activation thresholds to Bre1 (Rnf20/Rnf40) leads to replication stress and chromosomal generate progenitor patterning in the developing CNS. Development instability. Cancer Res 2012;72:2111–9. 2011;138:3711–21. 29. Northcott PA, Shih DJ, Peacock J, Garzia L, Morrissy AS, Zichner T, 46. Merchant AA, Matsui W. Targeting Hedgehog—a cancer stem cell et al. Subgroup-specific structural variation across 1,000 medullo- pathway. Clin Cancer Res 2010;16:3130–40. blastoma genomes. Nature 2012;488:49–56. 47. Barretina J, Caponigro G, Stransky N, Venkatesan K, Margolin AA, 30. Sasaki H, Hui C, Nakafuku M, Kondoh H. A binding site for Gli proteins Kim S, et al. The Cancer Cell Line Encyclopedia enables predictive is essential for HNF-3b floor plate enhancer activity in transgenics and modelling of anticancer drug sensitivity. Nature 2012;483:603–7. can respond to Shh in vitro. Development 1997;124:1313–22. 48. Bermudez O, Hennen E, Koch I, Lindner M, Eickelberg O. Gli1 mediates 31. Kimura H, Stephen D, Joyner A, Curran T. Gli1 is important for medul- lung cancer cell proliferation and Sonic Hedgehog–dependent mes- loblastoma formation in Ptc1þ/ mice. Oncogene 2005;24:4026–36. enchymal cell activation. PLoS ONE 2013;8:e63226. 32. Rodier F, Campisi J. Four faces of cellular senescence. J Cell Biol 49. Łastowska M, Al-Afghani H, Al-Balool HH, Sheth H, Mercer E, Coxhead 2011;192:547–56. JM, et al. Identification of a neuronal transcription factor network 33. Pasca di Magliano M, Hebrok M. Hedgehog signalling in cancer involved in medulloblastoma development. Acta Neuropathol Com- formation and maintenance. Nat Rev Cancer 2003;3:903–11. mun 2013;1:35. 34. Goodrich LV, Milenkovic L, Higgins KM, Scott MP. Altered neural cell 50. Schwartzentruber J, Korshunov A, Liu XY, Jones DT, Pfaff E, Jacob K, fates and medulloblastoma in mouse patched mutants. Science et al. Driver mutations in histone H3.3 and chromatin remodelling 1997;277:1109–13. genes in paediatric glioblastoma. Nature 2012;482:226–31.

www.aacrjournals.org Cancer Res; 74(8) April 15, 2014 OF11

Identification of FoxR2 as an Oncogene in Medulloblastoma

Hideto Koso, Asano Tsuhako, Eli Lyons, et al.

Cancer Res Published OnlineFirst March 5, 2014.

Updated version Access the most recent version of this article at: doi:10.1158/0008-5472.CAN-13-1523

Supplementary Access the most recent supplemental material at: Material http://cancerres.aacrjournals.org/content/suppl/2014/03/05/0008-5472.CAN-13-1523.DC1

E-mail alerts Sign up to receive free email-alerts related to this article or journal.

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Subscriptions Department at [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://cancerres.aacrjournals.org/content/early/2014/03/31/0008-5472.CAN-13-1523. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.