Oncogenic YAP Promotes Radioresistance and Genomic Instability in Medulloblastoma Through IGF2-Mediated Akt Activation

Oncogene (2012) 31, 1923–1937 & 2012 Macmillan Publishers Limited All rights reserved 0950-9232/12 www.nature.com/onc ORIGINAL ARTICLE Oncogenic YAP promotes radioresistance and genomic instability in medulloblastoma through IGF2-mediated Akt activation

A Fernandez-L1, M Squatrito1, P Northcott2, A Awan3, EC Holland1, MD Taylor2, Z Nahle´3 and AM Kenney3

1Department of Cancer Biology and Genetics, Memorial Sloan-Kettering Cancer Center, New York, NY, USA; 2Program in Developmental and Stem cell Biology, Division of Neurosurgery, Arthur and Sonia Labatt Brain Tumor Research Center, Hospital for Sick Children, University of Toronto, Toronto, Ontario, Canada and 3Departments of Neurosurgery and Cancer Biology, Vanderbilt University, Nashville, TN, USA

Radiation therapy remains the standard of care for many remains relatively high. Moreover, survivors suffer from cancers, including the malignant pediatric brain tumor serious long-term treatment-related sequelae. Current medulloblastoma. Radiation leads to long-term side therapy combines surgery and chemotherapy with effects, whereas radioresistance contributes to tumor craniospinal irradiation. However, the long-term side recurrence. Radio-resistant medulloblastoma cells occupy effects of radiation therapy as well as the potential the perivascular niche. They express Yes-associated induction of secondary malignancies have inspired many protein (YAP), a Sonic hedgehog (Shh) target markedly investigators to find ways to reduce the amount of elevated in Shh-driven medulloblastomas. Here we report radiation used (Mueller and Chang, 2009). that YAP accelerates tumor growth and confers Medulloblastomas bearing evidence of Sonic hedge- radioresistance, promoting ongoing proliferation after hog (Shh) pathway activity, approximately 30% of cases radiation. YAP activity enables cells to enter mitosis with (Northcott et al., 2011), are proposed to arise from un-repaired DNA through driving insulin-like growth cerebellar granule neuron precursors (CGNPs) (Eber- factor 2 (IGF2) expression and Akt activation, resulting hart, 2008). During cerebellar development, these cells in ATM/Chk2 inactivation and abrogation of cell cycle undergo a rapid expansion phase, and their proliferation checkpoints. Our results establish a central role for YAP depends on activation of the Shh signaling pathway in counteracting radiation-based therapies and driving (Dahmane and Ruiz i Altaba, 1999; Wallace, 1999; genomic instability, and indicate the YAP/IGF2/Akt axis Wechsler-Reya and Scott, 1999). Shh, a secreted ligand, as a therapeutic target in medulloblastoma. signals by binding to the transmembrane protein Patched Oncogene (2012) 31, 1923–1937; doi:10.1038/onc.2011.379; (Ptc), a tumor suppressor whose function is to inhibit the published online 29 August 2011 activity of Smoothened, a transmembrane protein resembling a G-protein-coupled receptor. In CGNPs, Keywords: medulloblastoma; sonic hedgehog; cerebellum; Shh interactions with Ptc relieve the inhibition of YAP;IGF;DNAdamage Smoothened and permit downstream signaling activity, resulting in the activation of Gli, N-myc and E2F1 transcription factors, and stabilization of insulin receptor substrate-1, all of which contribute to proliferation (Ruiz Introduction i Altaba, 1999; Oliver et al., 2003; Kenney et al., 2004; Parathath et al., 2008; Bhatia et al., 2011). Loss of Ptc or Medulloblastoma, a primitive neuro-ectodermal tumor, activating mutations in Smoothened are found in human is the most common malignant solid tumor of child- medulloblastomas, and can be phenocopied in mice, hood. These tumors arise in the cerebellum, a brain through mutational inactivation of Ptc or transgenic structure that regulates posture and coordination, and expression of activating alleles of Smoothened (Neu- develops largely perinatally. Current medulloblastoma roD2-SmoA1) in the cerebellum (Goodrich et al., 1997; therapy entails surgery, radiation and chemotherapy. Hatton et al., 2008). Recent observations in mouse Over the past half century, the long-term survival rate models have demonstrated the importance of genome for children with medulloblastoma has improved surveillance, as loss of DNA damage repair mechanisms remarkably ranging from 40 to 70% after 5 years, yet can lead to genomic instability in neural progenitor cells, recurrence is frequent and the overall mortality rate resulting in medulloblastoma, and can also cooperate with Shh signaling to enhance medulloblastoma forma- Correspondence: Dr AM Kenney, Departments of Neurosurgery and tion in mice (Pazzaglia et al., 2006; Leonard et al., 2008; Cancer Biology, Vanderbilt University, T-4224 MCN, Nashville, TN Tanori et al., 2008; Frappart et al., 2009). 37232, USA. Recently, we have shown that Shh induces expression E-mail: [email protected] Received 7 February 2011; revised 6 July 2011; accepted 26 July 2011; and activity of Yes-associated protein (YAP), a trans- published online 29 August 2011 criptional co-activator that is the target of the tumor- YAP promotes radioresistance in medulloblastoma A Fernandez-L et al 1924 suppressive Hippo pathway, and that YAP is highly vessels, a micro-environment known as the perivascular expressed in Shh-associated medulloblastomas. YAP is niche (PVN). These cells are proposed to have ‘tumor- known to have oncogenic properties, including the repopulating’ properties, and indeed, after irradiation, ability to promote epithelial–mesenchymal transition, these YAP-expressing cells can be found proliferating, suppression of apoptosis, growth factor-independent possibly contributing to the re-growth of the tumor. proliferation and anchorage-independent growth in soft To determine whether YAP expression increases MBC agar (Overholtzer et al., 2006). YAP is involved in the tumorigenicity, we cultured cells from medulloblasto- formation of numerous tumors, including glioblasto- mas arising in NeuroD2-SmoA1 mice, infected them mas, oral squamous-cell carcinomas, and cancers of the with retroviruses carrying either YAP or green fluor- pancreas, lung, ovary and cervix (Zeng and Hong, 2008; escent protein (GFP), and then implanted these cells Zhao et al., 2008; Fernandez-L and Kenney, 2010). intra-cranially into postnatal day 2 non-obese diabetic/ However, the effects exerted by YAP are highly context- severe combined immune deficiency pups. We then dependent, as it acts as a tumor suppressor in other monitored the animals for symptomatic evidence of malignancies, such as some breast cancers (Yuan et al., medulloblastoma (head tilt, seizures), confirmed upon 2008). Similarly, YAP activity can have pro- or being killed and dissection of the tumors. Ectopic antiapoptotic effects depending on the tissue and cell expression of YAP enhances medulloblastoma growth, context (Bertini et al., 2009). as determined by the ability of YAP-infected NeuroD2- Here, we show that YAP contributes to medulloblas- SmoA1 cells to form tumors in the recipient mice, more toma growth and medulloblastoma cell (MBC) survival effectively than GFP-infected NeuroD2-SmoA1 cells after irradiation, through driving expression and (14/17 YAP-SmoA1-injected mice vs 8/17 GFP-SmoA1- production of insulin-like growth factor 2 (IGF2) and injected mice). A similar trend was observed when wild- subsequent activation of Akt. Akt activity endows type C57BL/6 pups were used as recipients (Supple- tumor cells with the ability to evade the G1/S and mentary Figure 1). Moreover, YAP expression results G2/M checkpoints after irradiation, promoting in a significant acceleration of lethality (long-rank their ongoing proliferation despite the presence of test, P ¼ 0.0047; Figure 1a; P ¼ 0.0487; Supplementary un-repaired DNA, thereby contributing to genomic Figure 1), suggesting that YAP-expressing medullo- instability. These novel findings reveal a link between blastomas are more aggressive. YAP-SmoA1 medullo- oncogenic hedgehog signaling and genomic instability, blastomas have increased proliferation as determined and they suggest that targeting YAP- and IGF-driven by increased levels of cyclin D2 and phosphorylated signaling pathways may have therapeutic value by histone H3 (P-Hist3), markers of G1/S transition and restoring radiosensitivity and also permitting reduction mitosis phase of the cell cycle, respectively (Figures 1b of the radiation dose required to induce medulloblasto- and c). YAP-SmoA1 tumors also showed reduced levels ma tumor cell death. of apoptosis as determined by decreased levels of cleaved caspase 3 compared with tumors arising from GFP-infected SmoA1 cells. In keeping with our previous Results results, YAP is found throughout the tumor, but is most highly expressed in PVN cells. Interestingly, YAP- YAP expression promotes MBC tumorigenicity expressing SmoA1 medulloblastomas also bear evidence We have previously shown that YAP is highly expressed of increased angiogenesis, suggested by higher levels of in both human and mouse medulloblastomas, the YAP vascular endothelial growth factor and the endothelial locus is highly amplified in 3% of the human medullo- cell marker CD31 (Figures 1b and c). blastomas analyzed and YAP expression is sufficient to We have previously shown that YAP-expressing cells drive proliferation of CGNPs in the absence of their in Shh-induced mouse medulloblastomas proliferate obligate mitogen Shh (Fernandez-L et al., 2009). In after the animals have been treated with radiation. We NeuroD2-SmoA1 mouse medulloblastomas, YAP is next wished to determine whether YAP expression most highly expressed in tumor cells surrounding blood promotes survival after irradiation. To this end, we

Figure 1 YAP promotes medulloblastoma cell tumorigenicity, proliferation and survival. (a) Tumor-free survival curve. Two hundred thousand NeuroD2-SmoA1 tumor cells transduced with either GFP or YAP were implanted into the brain of non-obese diabetic/ severe combined immune deficiency postnatal day 2 pups. Animals were killed when they developed symptoms. YAP-overexpressing medulloblastomas occurred with a higher penetrance and earlier development compared to GFP-transduced tumors (log-rank test, P ¼ 0.0047). In all, 14/17 animals implanted with YAP-SmoA1 medulloblastoma cells developed tumors, whereas only 8/17 animals implanted with GFP-SmoA1 cells did. A similar trend was observed when non-immunocompromised recipient pups were used (log- rank test, P ¼ 0.0487; Supplementary Figure 1). (b) Representative western blot analysis of GFP-SmoA1 and YAP-SmoA1 medulloblastomas. Note reduced cleaved caspase 3, increased cyclin D2 and increased vascular endothelial growth factor in YAP- SmoA1 medulloblastoma. (c) Immunohistochemical analysis of YAP, the endothelial cell marker CD31, cleaved caspase 3 and phosphorylated histone H3 in GFP-SmoA1 (top) and YAP-SmoA1 (bottom) medulloblastomas. YAP-SmoA1 tumors have significantly reduced apoptosis and increased proliferation compared with GFP-SmoA1 tumors, indicated in the graphs. (d) Immunohistochemical analysis and quantification of apoptosis (cleaved caspase 3) and proliferation (phosphorylated histone H3) in GFP-SmoA1 and YAP-SmoA1 medulloblastomas 3 h after tumor-bearing animals were irradiated (2 Gy). Statistically significant differences are indicated as *Po0.05; ***Po0.001.

Oncogene YAP promotes radioresistance in medulloblastoma A Fernandez-L et al 1925 treated GFP-SmoA1 and YAP-SmoA1 medulloblasto- irradiation. We observed a marked induction of cleaved ma-bearing mice with 2 Gray (Gy) ionizing radiation caspase 3 in both tumors, but the levels of cleaved and analyzed cell death and proliferation 3 h post- caspase 3 were less in YAP-SmoA1 tumors as compared

Survival Curve SmoGFP SmoYAP 100 MB MB

80 GFP YAP1 60 Cl.Casp3

40 YAP CycD2

Percent survival 20 p=0.0047 VEGF 0 Tub 0 50 100 150 200 250 Time

YAP CD31 Cl.Casp3 P-Hist3

SmoGFP MB

SmoYAP MB

20x 20x 100x 100x

Cl.Casp3 staining P-Hist3 staining quantification quantification 1.5 2.5 * 2.0 1.0 * 1.5 1.0 0.5 0.5 0.0 0.0 Relative area stained Relative area stained GFP YAP GFP YAP

IR Cl.Casp3 staining quantification SmoGFP MB SmoYAP MB 1.5 Cleav. Casp3 1.0 *** 0.5

0.0 Relative area stained 100x GFP YAP P-Hist3 staining P-Hist3 2.5 quantification 2.0 * 1.5 1.0 0.5 100x 0.0 Relative area stained GFP YAP

Oncogene YAP promotes radioresistance in medulloblastoma A Fernandez-L et al 1926 to GFP-SmoA1 tumors (Figure 1d). Moreover, irra- Ionizing radiation causes the formation of double- diated YAP-SmoA1 medulloblastomas had significantly strand DNA breaks that are responsible for the higher levels of P-Hist3 and Ki67 (Supplementary activation of the DNA damage response (DDR), a Figure 1B) in comparison with irradiated GFP-SmoA1 complex network of proteins required for cell-cycle medulloblastomas. These lines of evidence indicate checkpoint maintenance and DNA repair. The G1/S that YAP expression endows MBCs with increased checkpoint prevents cells from entering the S phase, proliferation and survival capacities, and renders them whereas the G2/M checkpoint arrests cells with radio-resistant. un-repaired DNA before entry into mitosis. We wanted to assess whether high YAP expression causes defects in YAP expression promotes increased proliferation, survival cell cycle checkpoints, allowing cells to go through the and cell cycle checkpoints over-ride in irradiated cell cycle regardless of damage to their DNA. To cerebellar neural precursors examine the G1/S checkpoint, we exposed CGNPs to CGNPs are proposed to be the cell-of-origin for ionizing radiation (10 Gy), and determined the percen- medulloblastomas associated with increased Shh path- tage of cells in the S phase 18 h post-treatment by way activity (Wechsler-Reya and Scott, 1999). Consis- quantifying bromo-deoxyuridine (BrdU) incorporation tent with this hypothesis, treatment of CGNPs in culture during a 1-h pulse. Upon IR treatment, both GFP- and with exogenous Shh causes expression of many of the YAP-expressing CGNPs showed a strong reduction of same proliferation markers seen in this class of the number of cells incorporating BrdU; however, YAP- medulloblastomas, including Gli1, N-myc and YAP. expressing cells have a significantly higher S-phase ratio We wished to determine whether we could use in vitro (%BrdU-positive of treated/%BrdU-positive of un- primary CGNP cultures to dissect mechanistically the treated), suggesting a mild defect in the G1/S checkpoint effects of YAP on proliferation and survival after (Figure 2c). The infection efficiency was similar in both irradiation. Cells prepared from postnatal days 4 to 5 GFP- and YAP-infected cultures (representative images; mouse cerebella are maintained in the serum-free Supplementary Figure 2). To determine whether the G2/ medium; addition of Shh sustains their proliferation. M checkpoint was disabled in YAP-infected irradiated We transduced Shh-treated, proliferating CGNPs with CGNPs, we evaluated P-Hist3 levels at an early time retroviruses carrying GFP or YAP, and then irradiated point (1 h) after radiation. GFP-transduced cells showed them (10 Gy) 24 h later. As shown in Figure 2a, ectopic an 80% reduction of the number of P-Hist3-positive YAP expression is associated with increased CGNP cells as compared to untreated cells. In contrast, YAP- proliferation as determined by cyclin D2 levels, and expressing CGNPs showed only a 50% reduction of as we have reported previously (Fernandez-L et al., cells undergoing mitosis, indicating that the G2/M 2009). At 20 h post-irradiation, YAP-infected CGNPs checkpoint was defective (Figure 2c). had reduced levels of cleaved caspase 3 in comparison with GFP-infected CGNPs, and reduced numbers of apoptotic cells as determined by quantification YAP overexpression affects DNA repair and causes of pyknotic nuclei (GFP: 62.5±0.5% vs YAP: genomic instability 35.5±5.5%). YAP-transduced CGNPs were signifi- Defects in cell cycle checkpoints might result in genomic cantly more proliferative, as indicated by levels of cyclin instability. We asked at which level YAP was affecting D2, and by Ki67 and P-Hist3 staining (Figures 2a and the DDR. Breaks in DNA are marked by the formation b). Taken together, these results indicate that CGNP of protein complexes termed ionizing radiation-induced cultures recapitulate the in vivo effect of YAP expression foci, which can be identified when analyzed by on promoting survival and proliferation after irradia- immunostaining for proteins comprising these com- tion. Moreover, they suggest that YAP-mediated radia- plexes. To determine whether YAP-infected cells had tion resistance may contribute to medulloblastoma reduced DNA damage, we carried out immunostaining recurrence by promoting increased tumor cell survival for p53-binding protein 1 (p53BP1), an important and proliferation. mediator of the DDR often used as a marker of the

Figure 2 YAP promotes proliferation and survival in irradiated CGNPs. (a) Western blot analysis of proliferation (cyclin D2) and apoptosis (cleaved caspase 3) in GFP- or YAP-transduced Shh-treated CGNPs untreated (NT) and 20 h after g-irradiation (IR) with 10 Gy. Panel to the right shows overexposed image. Similar infection efficiencies were obtained with both retroviruses (Supplementary Figure 2). (b) Immunofluorescence analysis and quantification of cleaved caspase 3 and the proliferation markers Ki67 and phosphorylated histone H3 (P-Hist3) in GFP- or YAP-transduced CGNPs 20 h post-irradiation. (c) Left, immunofluorescence analysis of BrdU incorporation (1-h pulse) untreated (NT; top panels) and 18 h after irradiation (IR; middle panels). The S-phase ratios for GFP- and YAP-transduced CGNPs (%BrdU-positive of treated/%BrdU-positive of untreated) are shown at the bottom. Irradiated GFP-transduced cells were normalized to non-irradiated GFP-transduced CGNPs; irradiated YAP-transduced CGNPs were normalized to non-irradiated YAP-transduced CGNPs. The higher S-phase ratio indicates that YAP-overexpressing CGNPs present a defect in the G1/S checkpoint. Right, immunofluorescence analysis of phosphorylated histone H3 before (NT; top panels) and 1 h after radiation (IR; middle panels) in GFP- or YAP-transduced CGNPs. YAP-expressing CGNPs showed a higher mitotic ratio (%P-Hist3- positive of treated/% P-Hist3-positive of untreated) (bottom right), indicating failure of the G2/M checkpoint in these cells. Again, irradiated GFP-transduced cells were normalized to non-irradiated GFP-transduced CGNPs; irradiated YAP-transduced CGNPs were normalized to non-irradiated YAP-transduced CGNPs. Statistically significant differences are indicated as *Po0.05; ***Po0.001.

Oncogene YAP promotes radioresistance in medulloblastoma A Fernandez-L et al 1927 ionizing radiation-induced foci. At 3 h post-irradiation, cantly reduced number of cells containing foci as well GFP- and YAP-infected CGNPs had similar numbers of as fewer foci per cell (Figure 3a). These ﬁndings foci per cell (data not shown). However, at 9 and 24 h would suggest that YAP expression either enables cells post-irradiation, YAP-infected CGNPs had a signiﬁ- to repair DNA more rapidly, causing foci to resolve

NT IR IR GFP YAP GFP YAP GFP YAP

YAP1 YAP1

CycD2 CycD2

ClCasp3

Tub

GFP YAP Cl.Casp3 staining quantification Cl. Casp3 1.5

1.0 *

0.5 of staining

Relative Intensity 0.0 GFP YAP

Ki67 staining Ki67 quantification 5 *** 4 3 2

of staining 1

Relative Intensity 0 GFP YAP P-Hist3 P-Hist3 staining quantification 5 4 *** 3 2

of staining 1

Relative Intensity 0 GFP YAP

GFPYAP GFP YAP BrdU P-Hist3

BrdU positive cells P-Hist3 positive cells 0.0020 0.6 *** *** 0.0015 0.4 0.0010 0.2 0.0005 Mitosis Ratio S-phase Ratio 0.0000 0.0 GFP YAP GFP YAP

Oncogene YAP promotes radioresistance in medulloblastoma A Fernandez-L et al 1928 NT GF YA P P DNA damage-induced foci *** *** 100 IR+9h

50 % Cells

0 IR+24h GFP YAP GFP YAP GFP YAP NT IR+9h IR+24h Foci

No Foi 4.57 3.36 2.72 2.1 foci/cell foci/cell foci/cell foci/cell

IR+9h IR+24h H2O2 GFP YAP H2O2 GFP YAP IR+9h 120 100 *** 80 E 60 40

Tail length 20 0 NT NT NT GFP YAP H202 IR+24h 100 *** 80 60 40

Tail length 20 0 NT GFP YAP H202

** GFP YAP 30

Number of cells 0 GFP YAP Breaks No breaks

Number of observed breaks/cell Total 01234567 8 cells

Number GFP 41 4 2 0 1 0 0 1 1 50 of cells YAP 22 8 12 3 2 2 1 0 0 50

Figure 3 YAP overexpression prevents complete DNA repair. (a) Immunostaining (left) for p53BP1, which marks sites of DNA damage (‘foci’) in control (NT) or irradiated (IR) CGNPs infected with either GFP or YAP. Quantification shown at right indicates that fewer YAP-transduced cells contained foci compared to GFP-transduced CGNPs, and YAP-transduced cells contained fewer foci/ cell. (b) Comet assay to detect DNA damage performed in CGNPs 9 and 24 h post-irradiation. H2O2 serves as a positive control for detection of damaged DNA by this assay. YAP-overexpressing cells contained more damaged DNA at both time points, as measured by the length of the Comets’ tails (quantification at right). (c) (Left) Quantification of chromatid and chromosome breaks in metaphase spreads from GFP- or YAP-transduced CGNPs 24 h after irradiation. Right, representative image of metaphase spreads from GFP- or YAP-transduced irradiated CGNPs; arrows show examples of broken chromosomes. Table summarizes the number of breaks/cell after analysis of 50 irradiated GFP- or YAP-transduced CGNPs. YAP transduction was also associated with increased genomic instability in a mouse medulloblastoma cell line (Supplementary Figure 3B). Statistically significant differences are indicated as **Po0.01; ***Po0.001. earlier, or that YAP expression inhibits DNA repair To distinguish between these two possibilities, we pathways, resulting in dismantling of the DNA repair carried out comet assays to detect broken DNA, in complexes. irradiated GFP- or YAP-infected CGNPs. In the comet

Oncogene YAP promotes radioresistance in medulloblastoma A Fernandez-L et al 1929 assay, cells are immersed in an agarose matrix extended YAP-expressing irradiated CGNPs enter mitosis despite on a microscope slide, and treated with a lysis solution. the presence of double-stranded DNA breaks, and they Nuclear DNA is unwound under alkaline conditions suggest that the G2/M arrest checkpoint is compromised and fragments resulting from strand breaks can migrate by YAP expression. away from the nuclei under an electric field, forming a These observations prompted us to examine whether ‘comet tail’ when stained. We quantified the amount of YAP expression blocked activation of the G2/M DNA damage by measuring tail lengths. As shown in checkpoint by affecting the activity of proteins involved Figure 3b, comet tails can be detected in GFP-infected in its regulation. Upon irradiation, the kinase ATM is CGNPs at 9 h post-irradiation, but to a lesser extent at activated by autophosphorylation and subsequently 24 h post-irradiation, indicating that most of the phosphorylates and activates Chk2 kinase. Chk2, in damaged DNA is repaired at that point. turn, phosphorylates cdc25c, resulting in its inactivation In contrast, at 9 h after irradiation, YAP-infected cells and preventing entry into mitosis (Reinhardt and had strikingly pronounced comet tails. Comet tails Yaffe, 2009). Irradiation resulted in increased levels of persisted in YAP-infected CGNPs 24 h after irradiation, phosphorylated ATM and its substrate Chk2 in both a time point at which we can detect proliferation in these GFP- and YAP-infected cells. However, in the presence cells (Figure 2). To analyze whether the DNA repair of ectopic YAP, ATM and Chk2 were more rapidly defect leads to chromosomal alterations, we performed dephosphorylated (Figure 4b; western blot with quanti- metaphase spreads on GFP- or YAP-infected CGNPs fication below). We observed a similar trend towards 24 h after irradiation. As shown in Figure 3c, YAP Chk2 inactivation in MBCs obtained from SmoA1 infection is associated with an increase in the number of medulloblastomas, cultured in vitro and transduced cells with persistent DNA breaks. We also observed that with either GFP or YAP. Our observation of ATM YAP-transduced Pzp53med cells (Berman et al., 2002) and Chk2 inactivation in YAP-infected CGNPs and showed a similar loss of irradiation-induced foci and MBC is consistent with relief from the G2/M check- increased the number of cells with DNA breaks after point, and consequently, re-entry into mitosis despite irradiation (Supplementary Figure 3A and B). These the presence of un-repaired DNA breaks. We also observations indicate that YAP expression enables cells observed reduced phosphorylation of histone H2AX in to proliferate with un-repaired DNA, a condition that YAP-transduced irradiated CGNPs, further indicative can lead to genomic instability, a hallmark of cancer of impaired DDR pathway signaling (Supplementary cells. To address whether the effect of YAP was specific Figure 4A). Interestingly, we did not observe differential for radiation-induced DNA damage, we treated GFP- effects of YAP expression on the regulation of Chk1 or and YAP-infected cells with the antibiotic bleomycin, p53 activity after exposure to radiation in CGNPs or which causes double-strand breaks in DNA. Similar to MBCs, indicating that YAP’s downstream effectors what we observed in irradiated cells, 9 and 24 h after preferentially target the ATM/Chk2 DDR pathway treatment there were significantly fewer cells that (Figure 4b). contained foci among those overexpressing YAP, which indicates that the role of YAP promoting focus IGF2 is a candidate YAP target mediating proliferation disassembly is not specific to radiation-induced damage and survival after irradiation in the DNA (Supplementary Figure 3C). We next wished to gain insight as to the mechanism through which YAP protects tumor cells from radiation- induced apoptosis and allows them to continue cycling. YAP expression promotes inactivation of the checkpoint YAP functions in a complex with TEAD to regulate regulators ATM and Chk2 gene expression and proliferation (Zhao et al., 2008). To Entry into mitosis is marked by activation of the cyclin- determine which genes may be induced or suppressed by dependent kinase (Cdk)1:cyclin B1 complex. During G2, YAP in irradiated CGNPs, we carried out microarray Cdk1 is maintained in an inactive state by phosphoryla- analysis of mRNA prepared from GFP- or YAP- tion on Thr 14 and Tyr 15 (Castedo et al., 2002); de- infected control or irradiated CGNPs. Among the genes phosphorylation by cdc25 disinhibits Cdk1. To confirm most highly expressed in YAP-infected CGNPs was that YAP-expressing cells continued to enter mitosis IGF2, confirmed by quantitative reverse transcription– after irradiation, indicating inactivation of the G2/M polymerase chain reaction (PCR) (Figure 5a). IGF2 checkpoint, we carried out western blotting for Tyr 15- mRNA expression was increased in both cell popula- phosphorylated Cdk1. As shown in Figure 4a, irradia- tions after irradiation, but the levels of IGF2 in the tion results in increased Tyr15-phosphorylated Cdk1 in YAP-infected CGNPs were significantly higher than both GFP- and YAP-infected CGNPs. However, after that in the GFP-infected cell under either condition 3 h, YAP-infected CGNPs show reduced levels of Tyr15- (non-irradiated or irradiated). When we analyzed IGF2 phosphorylated Cdk1, indicating re-activation of Cdk1 protein levels in GFP- or YAP-infected CGNPs, we at this early time point, when DNA breaks still persist. found that YAP-infected CGNPs expressed and secreted The reduction of inactive Cdk1 was associated with an significantly more IGF2 (Figures 5b and c). increase in levels of cyclin B1, which is synthesized IGF signaling is required for survival of CGNPs during interphase. Following metaphase, cyclin B1 is in vivo and in vitro, and increased activity of the IGF degraded; this degradation is required for completion pathway is found in human medulloblastomas (Hahn of mitosis. Taken together, these results confirm that et al., 2000; Chrysis et al., 2001; Rao et al., 2004;

Oncogene YAP promotes radioresistance in medulloblastoma A Fernandez-L et al 1930 NT IR+1h IR+3h GFP YAP GFP YAP GFP YAP P-Cdk1 Y15

CycB1

Tub

CGNPs MBCs NT IR+1h IR+3h IR+6h NT IR+1h IR+3h GFP YAP GFP YAP GFP YAP GFP YAP GFP YAP GFP YAP GFP YAP

YAP

P-ATM

P-Chk2

Chk2

P-Chk1

Chk1

P-p53

p53

Tub

CGNPs MBCs

ATM phosphorylation Chk2 phosphorylation ATM phosphorylation upon Chk2 phosphorylation upon upon radiation upon radiation radiation radiation 1500 30 25 50 GFP GFP GFP GFP YAP YAP 20 YAP 40 YAP 1000 20 15 30 10 20 500 10 Relative units

Relative units 5 10 Relative units Relative units 0 0 0 0 No γ γ+1h γ+3h γ+6h No γγ+1h γ+3h γ+6h No γ γ+1h γ+3h No γ γ+1h γ+3h Time point Time point Time point Time point Figure 4 The ATM-Chk2 DDR pathway is rapidly deactivated in YAP-overexpressing cells. (a) Representative western blot analysis of GFP- or YAP-transduced CGNPs at the indicated time points after irradiation. CGNPs overexpressing YAP at 1 and 3 h post- irradiation showed reduced levels of inhibited (Tyr-15-phosphorylated) Cdk1, and higher levels of cyclin B1 indicating increased entry into mitosis. (b) Western blot analysis of GFP- or YAP-transduced CGNPs (left) or NeuroD2-SmoA1 cells (MBCs) (right) at the indicated times after irradiation. YAP-transduced CGNPs show lower levels of phosphorylated ATM and Chk2 3 h after radiation compared to GFP-transduced cells. Similar results were observed in irradiated MBCs transduced with either GFP or YAP. Quantiﬁcation of phosphorylated ATM and Chk2 levels are shown at the bottom. YAP expression did not differentially affect Chk1 or p53 levels or phosphorylation.

Hartmann et al., 2005; Corcoran et al., 2008; Tanori YAP-expressing tumors was conserved in YAP-SmoA1 et al., 2010). Interestingly, when we queried a genetically medulloblastomas, as determined by western blot and characterized database of over 100 human medulloblas- immunohistochemical analysis (Figures 5e and f). As we tomas (Northcott et al., 2009b), we found that IGF2 was observed in CGNPs, IGF2 levels are increased in both most highly expressed in the subclass of medulloblasto- GFP-SmoA1 and YAP-SmoA1 medulloblastomas after mas associated with the activation of the Shh pathway irradiation, but are higher in YAP-SmoA1 medullo- (T-test, P ¼ 3.553E-14) (Figure 5d). This class of tumors blastomas. also expresses high levels of N-myc, Gli1 and miR17/92 IGF2 acts as a secreted ligand, binding to and (Pomeroy et al., 2002; Kool et al., 2008; Northcott activating the IGF1 receptor. The predominant down- et al., 2009b). The increased level of IGF2 protein in stream effector of the activated IGF1 receptor is the

Oncogene YAP promotes radioresistance in medulloblastoma A Fernandez-L et al 1931 IGF2 mRNA expression IGF-2 ELISA *** 25 * 1.4 *** *** 20 15 1.2 10 1.0 Fold change 5 Fold induction 0 0.8 GFP YAP GFP YAP GFP YAPGFP YAP NT IR NT IR

12 11 NT IR SmoGFP SmoYAP 10 GFP YAP GFP YAP MB MB 9 IGF2 IGF2 IGF2 8

Tub 7 Tub 6 5

Fetal Adult WNT SHH Group CGroup D Normal Medulloblastoma cerebellum NT IR SmoGFP MB SmoYAP MB SmoGFP MB SmoYAP MB

100X 100X 100X 100X

NT IR

SmoGFP MB SmoYAP MB SmoGFP MB SmoYAP MB P-Akt

YAP

merge

Figure 5 YAP induces expression of IGF2 and activation of Akt. (a) IGF2 was found to be upregulated in the presence of YAP using microarray analysis, confirmed by quantitative RT–PCR. Irradiation induces IGF2 in GFP- and YAP-infected cells, but levels are significantly higher in YAP-infected CGNPs 3 h post-irradiation. (b) Enzyme-linked immunosorbent assay analysis of IGF2 production in GFP- or YAP-transduced control or irradiated CGNPs (3 h). Statistically significant differences are indicated as *Po0.05; ***Po0.001. (c) Western blot analysis of IGF2 protein in GFP- or YAP-transduced control and irradiated CGNPs (3 h). (d) Box plot showing IGF2 mRNA expression obtained from exon array profiling of 103 primary human medulloblastomas, 9 fetal and 5 adult human control cerebella. IGF2 is highly expressed specifically in SHH-driven medulloblastomas (t-test, P ¼ 3.553E-14). IGF1 is not specifically elevated in human Shh-associated medulloblastomas (Supplementary Figure 4). (e) Western blot analysis of IGF2 protein in GFP-SmoA1 and YAP-SmoA1 mouse medulloblastomas. (f) Immunohistochemical staining for IGF2 in medulloblastomas from non-irradiated (left) and irradiated (3 h; right) GFP-SmoA1 and YAP-SmoA1 tumor-bearing mice. (g) Immunofluorescence analysis of Akt S473 phosphorylation and YAP protein in medulloblastomas from non-irradiated (left) and irradiated (3 h; right) GFP- SmoA1 and YAP-SmoA1 tumor-bearing mice.

Oncogene YAP promotes radioresistance in medulloblastoma A Fernandez-L et al 1932 kinase Akt, which plays multiple roles in survival and of YAP on DNA damage-dependent focus formation proliferation. We have previously shown that Akt after irradiation (Figure 6c), causing a significant rescue cooperates with Shh signaling to promote proliferation of focus formation as determined by immunofluorescent through stabilization of N-myc (Kenney et al., 2004). staining for p53BP1. Moreover, IGF2 knockdown Akt activity has been linked to abrogation of the G2/M impaired the effects of YAP expression on CGNP checkpoint after irradiation, through inactivation of survival after irradiation, as well as proliferation ATM/Chk2 (Kandel et al., 2002; Hirose et al., 2005). (Figure 6d). These results are consistent with IGF2 Consistent with increased IGF1 receptor activity in and its downstream effector Akt being necessary for the response to IGF2 secretion, YAP-Smo-driven medullo- ability of YAP to inactivate the G2/M checkpoint, blastomas exhibit higher levels of activated Akt (S473- permitting ongoing proliferation and enhancing survival phosphorylated), most notably in cells surrounding the of irradiated cells. vasculature (Figure 5g); interestingly, these cells also express the highest levels of YAP and they have been proposed to function as tumor-repopulating cells after irradiation (Hambardzumyan et al., 2008; Fernandez-L Discussion et al., 2009). Taken together, these observations raise the possibility that YAP-mediated IGF2 induction not only Amplification of the YAP gene locus as well as YAP promotes survival and proliferation through Akt protein overexpression has been reported in a wide activation, but may also affect the phosphorylation of spectrum of human and murine tumors. We have ATM/Chk2, resulting in disruption of the G2/M previously shown that YAP is a downstream target of checkpoint. mitogenic Shh signaling that promotes proliferation and inhibits apoptosis of cerebellar granule cell precursors and we showed that YAP is overexpressed and ampli- IGF2 is required for YAP-mediated G2/M arrest over- fied in the Shh-associated class of medulloblastomas ride and cell survival and proliferation after irradiation (Fernandez-L et al., 2009). Here, we report a new role for To address whether IGF2/Akt activity regulates the YAP in promoting IGF2/Akt-dependent radioresistance YAP-associated DDR defect after radiation, we used and ongoing proliferation in the presence of un-repaired Shh-treated CGNP cultures. Confirming our observa- DNA, illustrating a novel link between Shh signaling, tions (Figure 5), YAP infection is associated with YAP and the Akt pathway. We show that YAP activity increased Akt activity (Figure 6a) as determined by markedly induces IGF2 transcription, in the absence phosphorylation of S473. Treatment of YAP-infected and presence of radiation. IGF2, in turn, activates Akt CGNPs with the drug LY294002, which inhibits downstream of phosphoinositide 3-kinase (PI3K), pro- phosphoinositide-3 kinase, the upstream activator of viding cells with a survival and proliferative advantage, Akt, reduced levels of S473-phosphorylated Akt. After and allowing them to inactivate cell cycle checkpoints irradiation, we saw induction of ATM phosphorylation (Figure 7). Many studies have implicated PI3K and Akt in GFP- and YAP-infected CGNPs, although to a lesser in cell survival, through its effects on FKHR proteins extent in the presence of ectopic YAP, indicating and Bcl2 family members (Brunet et al., 2001; Burgering reduced activity of this kinase. In keeping with reduced and Kops, 2002; Duronio, 2008), and Akt is essential for ATM activity in the presence of YAP, we also observed CGNP survival (Dudek et al., 1997). Transcriptional reduced Chk2 phosphorylation. In the presence of profiling of human medulloblastomas has shown in- LY294002, full ATM and Chk2 phosphorylation were creased expression of IGF2 (Pomeroy et al., 2002), and recovered, indicating that YAP requires Akt activity for our report clearly links high levels of IGF2 expression its suppressive effect on ATM and Chk2. with the Shh-associated subclass of medulloblastomas. We next wished to determine whether IGF2 is IGF2 has been shown to be indispensable for the necessary for the effects of YAP on the DDR. To this formation and progression of Shh-driven murine me- end, we used retroviruses targeting IGF2 for short dulloblastomas (Hahn et al., 2000; Corcoran et al., hairpin RNA (shRNA)-mediated knockdown. As 2008) and to synergize with SHh in the induction of shown in Figure 6b, YAP-infected CGNPs transduced tumor formation (Rao et al., 2004). IGF2 has previously with these retroviruses showed strikingly reduced levels been shown to promote proliferation of MBCs and of IGF2, in comparison with YAP-infected CGNPs cerebellar neural cell precursors through the activation transduced with retroviruses carrying a scrambled, non- of Akt (Hartmann et al., 2005). specific shRNA sequence. Consistent with IGF2 being Our results show that ectopic YAP expression an upstream activator of Akt, IGF2 knockdown was accelerates medulloblastoma onset and aggressiveness, associated with reduced Akt Ser473 phosphorylation. In and also confers a survival advantage after addition, knocking down IGF2 in YAP-expressing cells tumor radiation. This has particularly important im- led to a recovery of phospho-ATM and phospho-Chk2 plications for medulloblastomas; owing to their radio- levels, as well as phospho-Cdk1, comparable to those sensitivity, radiation therapy is a standard in children observed in GFP-infected cells. The increase in inactive older than 3 years of age (Mueller and Chang, 2009). Cdk1 when IGF2 is knocked down indicates cells We show here that 3 h after radiation, YAP-driven arresting after radiation, due to a restored G2/M SmoA1 medulloblastomas feature higher levels of checkpoint. IGF2 knockdown also blocked the effects proliferation markers and reduced apoptosis. In vitro,

Oncogene YAP promotes radioresistance in medulloblastoma A Fernandez-L et al 1933 NT IR

shRNA : ScblIGF2 Scbl IGF2 NT IR GFPYAP YAP GFP YAP YAP LY - - + - - + GFP YAP YAP GFP YAP YAP YAP P-Akt IGF2

Akt P-Akt

P-ATM P-ATM

P-Chk2 P-Chk2

Tub P-Cdk1 Y15

Tub

Foci GFP + shScbl YAP + shScbl YAP + shIGF2 DNA damage-induced foci No Foci IR+9h ** ** *** ***

100

IR+24h % Cells

0 GFPYAPYAP GFP YAP YAP GFP YAP YAP shScbl shlGF2 shScbl shlGF2 shScbl shlGF2

NT IR+9h IR+24h

GFP + shScbl YAP + shScbl YAP + shIGF2

Cl.Casp3 Cl.Casp3 Cl.Casp3 CI.Casp3 staining quantification 1.5 ***

1.0

staining 0.5

GFP YAP YAP Relative intensity of 0.0 GFP YAP YAP shScbl shlGF2

Ki67 staining quantification 4 * * 3 Ki67 Ki67 Ki67 2

staining 1

Relative intensity of 0 GFP YAP YAP shScbl shlGF2

Figure 6 Akt inhibition or IGF2 knockdown abrogate the effects of YAP on DNA repair, proliferation and survival. (a) Western blot analysis of CGNPs infected with either GFP or YAP, treated with the PI3K inhibitor LY294002 or with vehicle, irradiated and analyzed 3 h after radiation. LY294002 treatment in YAP-expressing cells restored the levels of phosphorylated-Akt, -ATM and -Chk2 to the levels observed in GFP-expressing CGNPs. (b) Western blot showing effective knockdown of IGF2 in CGNPs infected with either GFP or YAP, and then transduced with lentiviruses targeting IGF2 for shRNA-mediated knockdown or encoding for a scrambled shRNA sequence. IGF2 knockdown reduces Akt phosphorylation and rescues ATM and Chk2 phosphorylation in YAP- transduced cells 3 h after irradiation. (c) Immunofluorescence analysis of p53BP1 localization to foci in irradiated GFP- or YAP- transduced CGNPs subsequently infected with retroviruses targeting IGF2 for shRNA-mediated knockdown or encoding for a scrambled shRNA sequence. Quantification is shown at right. Fewer YAP-transduced CGNPs contained foci 9 and 24 h post- irradiation compared to GFP-transduced CGNPs. IGF2 knockdown restored the percentage of cells containing DNA damage-induced foci to the levels observed in GFP-expressing CGNPs. (d) Immunofluorescence analysis of cleaved caspase 3 and Ki67 in GFP- or YAP-transduced CGNPs infected with either scrambled or IGF2-targeting retroviruses, 3 h after irradiation. IGF2 knockdown restored the levels of proliferation and apoptosis to those observed in GFP-expressing CGNPs. Statistically significant differences are indicated as *Po0.05; **Po0.01; ***Po0.001. we found a similar effect when we irradiated GFP- expressing cells after radiation, but also failed to or YAP-overexpressing CGNPs. CGNPs express- arrest in response to radiation, skipping the G2/M ing YAP not only proliferated more than GFP- checkpoint.

Oncogene YAP promotes radioresistance in medulloblastoma A Fernandez-L et al 1934 The ability of cells to maintain genomic integrity is IGF2 fundamental for protection from cancer development. YAP Genomic instability characterizes almost all sporadic human cancers, although mechanisms causing it are still not well understood. High-throughput sequencing studies suggest that mutations in DNA repair genes do not account for the presence of genomic instability in many sporadic cancers (Negrini et al., 2010). Alternatively, activation of

P oncogenes, and more generally of growth signaling path- Akt ways, has been shown to induce genomic instability in ATM mammalian cells cultured in vitro and in mouse models Chk2 (Hernando et al., 2004; Negrini et al., 2010). Hyperactive Shh pathway activity has been shown to induce genomic instability and the development of spontaneous and Survival DNA G1/S,G2/M ionizing radiation-induced tumors (Leonard et al., 2008). Proliferation repair checkpoints Now we show that high levels of YAP expression further contribute to the generation of genomic instability as irradiated YAP-expressing cells present a higher number of chromosomal alterations. The role of genomic instability has remained a controversial issue in cancer biology. Genomic Instability Although it is still unclear if it is necessary for tumorigenesis to occur, genomic instability certainly provides the tumor with an advantage in terms of faster progression through Figure 7 YAP-mediated IGF2 expression drives cell survival, the many stages of tumorigenesis (Sieber et al., 2003). proliferation and genomic instability. Model suggesting the Indeed, we show that ectopic expression of YAP accelerates mechanism through which oncogenic YAP expression permits cell medulloblastoma onset and produces tumors that are proliferation and survival after radiation-induced DNA damage. highly aggressive and rapidly lethal. YAP induces IGF2 expression in CGNPs and medulloblastoma YAP expression is particularly high in the PVN of cells. IGF2 activates the PI3K/Akt pathway, promoting survival and increasing proliferation of these cells. In addition, increased medulloblastomas, where the tumor-repopulating cells activation of Akt leads to faster ATM and Chk2 dephosphoryla- reside. We have previously shown that 48 h after radiation, tion after radiation, subsequently deactivating the DDR and the YAP-positive population starts to expand (Fernandez- removing the G1/S and G2/M checkpoints. Favoring survival, L et al., 2009). Hambardzumyan et al. (2008) elegantly proliferation and genomic instability confers an advantage to YAP- expressing tumor cells. Inhibiting IGF2/Akt signaling in tumors demonstrated that the PI3K-Akt pathway regulates with high YAP expression may prevent recurrence and enable use survival of these tumor-repopulating cells residing in the of lower radiation doses by increasing tumor cell radiosensitivity. PVN. These results point to the importance of YAP- regulating survival and proliferation post-radiation, allowing the cells to skip the cell cycle checkpoints and expand Defective G2/M checkpoint activation in the presence despite the un-repaired damage in their DNA, contributing of DNA damage has been linked to Akt activity. ultimately to genomic instability. Further studies will Activation of Akt suppresses the checkpoint by inhibit- address whether YAP-induced genomic instability is solely ing the DDR pathways, either ATM/Chk2 or ATR/ a consequence of the Akt-mediated cell cycle checkpoint Chk1, depending on the cellular context (Kandel et al., suppression or whether Akt lying downstream of YAP- 2002; Shtivelman et al., 2002; Hirose et al., 2005; Xu induced IGF2 could also inhibit the DNA repair et al., 2010). In agreement with these studies, we machinery, as previous studies have suggested (Plo et al., observed a decrease in phosphorylated ATM and 2008) (Figure 7). Our results suggest that targeting Akt Chk2 in irradiated YAP-overexpressing CGNPs and downstream of YAP would be useful for medulloblastoma MBCs compared to GFP-expressing cells, and YAP- therapy to reduce the dose of radiation needed and to transduced cells showed fewer DNA damage-induced prevent tumor recurrence, by radiosensitizing PVN-residing foci. The effects of YAP on ATM and Chk2 was tumor-repopulating cells. reversed in the presence of a PI3K inhibitor, and when we knocked down IGF2, which also abolished the accelerating effect of YAP on DNA damage-dependent Materials and methods focus disassembly. Interestingly, a previous study has shown that aberrant activation of the Shh pathway Animal studies through Ptc loss-of-function impairs the function of Harvest of CGNPs from neonatal mice, preparation of ATR/Chk1, a separate signaling cascade (Leonard et al., cerebella and tumor tissue from wild-type and mutant mice for histological analysis and irradiation of tumor-bearing mice 2008). We did not detect effects of YAP on this cascade were carried out in compliance with the Memorial Sloan- (data not shown); however, the impact we report of Kettering Institutional animal care and use committee guide- YAP on ATM/Chk2 may synergize with an already lines. NeuroD2-SmoA1 mice were provided by Jim Olson (Fred defective DDR due to Smoothened-mediated Chk1 Hutchinson Cancer Research Center, Seattle, MA, USA). impairment. Ptc þ /À mice were provided by Kathryn Anderson (Memorial

Oncogene YAP promotes radioresistance in medulloblastoma A Fernandez-L et al 1935 Sloan-Kettering Cancer Center, New York, NY, USA). Non- TaqMan probes used for quantitative PCR were IGF2 obese diabetic/severe combined immune deficiency mice were (Mm01163433-m1) and control HPRT1 (Mm01545399-m1). purchased from The Jackson Laboratory (Bar Harbor, ME, USA). Protein preparation and immunoblotting Protein extracts were prepared as described previously (Kenney Human tumor collection and expression analysis and Rowitch, 2000). Protein content was determined by using the Exon array profiling and data analysis with tumor subgroup- Bio-Rad protein assay. In all, 50–75 mg of each sample was ing were performed as published (Northcott et al., 2009a). separated by sodium dodecyl sulfate–polyacrylamide gel electro- phoresis on 8–10% polyacrylamide gels and then transferred in 20% methanol buffer at 4 1C to Immobilon polyvinylidene CGNPs and MBC culture difluoride (Millipore, Billerica, MA, USA) membranes. Immu- CGNP cultures were generated as described previously noblotting was carried out according to standard methods. (Kenney and Rowitch, 2000). Cells were plated on individual Antibodies used for western blotting were: YAP1 (Abcam, poly-DL-ornithine (Sigma-Aldrich, St Louis, MO, USA) pre- Cambridge, MA, USA), phospho-Akt (Ser473), phospho-Cdk1 coated plates or pre-coated glass coverslips. In all cases, cells (Tyr15), phospho-Chk2 (Thr68), total Akt and cleaved caspase 3 were treated with 3 mg/ml of Shh. Where indicated, LY294002 (Cell Signaling, Danvers, MA, USA), phospho-ATM (Ser1981) (Sigma-Aldrich) was used at a concentration of 10 nM for 6 h. (Rockland Immunochemicals, Inc., Gilbertsville, PA, USA), MBCs were obtained from SmoA1 mouse medulloblasto- Chk2 (Millipore), IGF2 and b-tubulin (Sigma-Aldrich), vascular mas. Briefly, tumors were disassociated and cells were endothelial growth factor, and cyclins D2 and B1 (Santa Cruz incubated in Trypsin/EDTA solution for 30 min, and then Biotechnology, Inc., Santa Cruz, CA, USA). Donkey anti-mouse passed through a cell strainer. Cells were subsequently horseradish peroxidase-linked secondary was from Jackson separated on a density step gradient of 35 and 60% Percoll Research Laboratories and goat anti-rabbit from Thermo solution (Sigma-Aldrich). Purified MBCs were enriched by pre- Scientific. Peroxidase activity was detected using Amershams’s plating on uncoated tissue culture dishes to remove adherent ECL reagents and exposing membranes to Kodak Biomax film. fibroblasts and glial cells. Non-adherent cells were plated on tissue culture dishes pre-coated with poly-D-lysine (Sigma- Aldrich) and Matrigel (BD Biosciences, Franklin Lakes, NJ, Cell and tissue immunostaining USA), infected with retroviruses 24 h later, and cultured for 24 For immunofluorescence, 8 mm sections of paraformaldehyde- or more hours before implantation or irradiation. fixed/paraffin-embedded tissues were first de-waxed and re- hydrated before antigen retrieval. CGNPs were cultured on poly-DL-ornithine-coated glass coverslips as described pre- Tumor cell implantation viously. Cells were fixed with 4% paraformaldehyde for In total, 0.2 million MBCs were injected in the cortices of P2 20 min. Immunofluorescent staining of sections and cells was non-obese diabetic/severe combined immune deficiency or carried out according to standard methods. Immunohistochem- C57BL/6 pups with a stereotaxic syringe. ical staining was performed using a Discovery XT automated staining processor (Ventana Medical Systems, Inc., Tucson, AZ, Retrovirus production and infection USA). Antibodies used were as follows: Ki67 (Vector Labs, The YAP cassette was cloned from pBabe-YAP1 (Addgene, Burlingame, CA, USA), YAP1 (Abcam), CD31 (BD Transduc- Cambridge, MA, USA) into the retroviral vector pWzl. tion Laboratories, Mississauga, ON, Canada), GFP (Invitro- shRNA interference sequences targeting IGF2 at the 30 or 50 gen), BrdU (BD Biosciences), p53BP1 and IGF2 (Novus end of the coding sequence were designed using the Cold Biologicals, Littleton, CO, USA), phospho-histone 3, cleaved Spring Harbor Laboratory RNAi OligoRetriever database and caspase 3 and phospho-Akt (Ser473) (Cell Signaling). ligated into pSHAG vector, and then transferred to MSCV retroviral vector and produced as described (Nahle et al., Image capturing 2008). A control scrambled shRNA retrovirus was also Staining of cultured primary cells and tissue sections was prepared. The 293 EBNA (Invitrogen, Carlsbad, CA, USA) visualized with a Leica DM5000B microscope and images were packaging cells were co-transfected with gag-pol and VSVg taken using the Leica FW400 software. For quantification, packaging plasmids plus pWzl-eGFP, pWzl-YAP1, pMSCV- TIFF images of 5–10 random fields were taken for each IGF2shRNA or a control scrambled shRNA using Fugene 6 experimental group and average pixel intensities were mea- transfection reagent (Roche Applied Science, Indianapolis, IN, sured using the ImageJ and Volocity softwares. USA). The media were changed 12 h after transfection and supernatants (8 ml) were harvested at 24 and 48 h and filtered through 0.45-mm syringe filters. For infection, Shh-treated Mitosis spreads CGNPs were exposed to the viral supernatants for 3 h. Viral Metaphase preparations were carried out by the Molecular supernatant was then removed and replaced with fresh Cytogenetics core facility at Memorial Sloan-Kettering Cancer medium þ Shh. Center according to standard procedures.

RNA extraction and real-time PCR Comet assay Total RNA from cerebellar granule neural precursors was The Comet assay was performed with the Trevigen kit and extracted and puriﬁed using the MiRvana kit (Ambion, Austin, following the manufacturer’s instructions. Tail lengths were TX, USA). cDNA was prepared from 1 mg total RNA by using calculated using the Comet Assay IV software. Averages from iScript cDNA Synthesis kit (Bio-Rad Laboratories, Hercules, four independent experiments were calculated and representa- CA, USA). Quantitative PCR was performed using TaqMan tive pictures are shown. Universal PCR Master Mix (Applied Biosystems, Foster City, CA, USA). RNA expression data were acquired and Enzyme-linked immunosorbent assay analyzed using a StepOnePlus Real-Time PCR system (Applied Supernatants from CGNPs infected with GFP or YAP and Biosystems). Average results and standard errors are presented. subsequently cultured in 800 ml minimal media without N2

Oncogene YAP promotes radioresistance in medulloblastoma A Fernandez-L et al 1936 supplement were collected. IGF2 enzyme-linked immuno- Acknowledgements sorbent assay was performed using the R&D systems kit and following the manufacturer’s instructions. Triplicates were We are grateful to Martine Roussel and her group at St Jude used in each case and the average from three independent Children’s Research Hospital for training and advice on MBC experiments is shown. culture and intracranial implantation. AF-L received support from the Spanish Ministry of Education, the Charles H Revson Foundation and the Childhood Brain Tumor Founda- tion. This work was supported by grants to AMK from the Conﬂict of Interest Alex’s Lemonade Stand Foundation, National Brain Tumor Society, James S McDonnell Foundation and the NIH The authors declare no conﬂict of interest. (NINDS R01NS061070).

References

Berman DM, Karhadkar SS, Hallahan AR, Pritchard JI, Eberhart the formation of medulloblastoma and rhabdomyosarcoma. J Biol CG, Watkins DN et al. (2002). Medulloblastoma growth inhibition Chem 275: 28341–28344. by hedgehog pathway blockade. Science 297: 1559–1561. Hambardzumyan D, Becher OJ, Rosenblum MK, Pandolfi PP, Bertini E, Oka T, Sudol M, Strano S, Blandino G. (2009). YAP: at the Manova-Todorova K, Holland EC. (2008). PI3K pathway regulates crossroad between transformation and tumor suppression. Cell survival of cancer stem cells residing in the perivascular niche Cycle 8: 49–57. following radiation in medulloblastoma in vivo. Genes Dev 22: Bhatia B, Hsieh M, Kenney AM, Nahle Z. (2011). Mitogenic Sonic 436–448. hedgehog signaling drives E2F1-dependent lipogenesis in progenitor Hartmann W, Koch A, Brune H, Waha A, Schuller U, Dani I et al. cells and medulloblastoma. Oncogene 30: 410–422. (2005). Insulin-like growth factor II is involved in the proliferation Brunet A, Datta SR, Greenberg ME. (2001). Transcription-dependent control of medulloblastoma and its cerebellar precursor cells. Am J and -independent control of neuronal survival by the PI3K-Akt Pathol 166: 1153–1162. signaling pathway. Curr Opin Neurobiol 11: 297–305. Hatton BA, Villavicencio EH, Tsuchiya KD, Pritchard JI, Ditzler S, Burgering BM, Kops GJ. (2002). Cell cycle and death control: long live Pullar B et al. (2008). The Smo/Smo model: hedgehog-induced Forkheads. Trends Biochem Sci 27: 352–360. medulloblastoma with 90% incidence and leptomeningeal spread. Castedo M, Roumier T, Blanco J, Ferri KF, Barretina J, Tintignac LA Cancer Res 68: 1768–1776. et al. (2002). Sequential involvement of Cdk1, mTOR and Hernando E, Nahle Z, Juan G, Diaz-Rodriguez E, Alaminos M, p53 in apoptosis induced by the HIV-1 envelope. EMBO J 21: Hemann M et al. (2004). Rb inactivation promotes genomic 4070–4080. instability by uncoupling cell cycle progression from mitotic control. Chrysis DC, Alexandrides TK, Koromantzou E, Georgopoulos N, Nature 430: 797–802. Vassilakos P, Kiess W et al. (2001). Novel application of IGF-I and Hirose Y, Katayama M, Mirzoeva OK, Berger MS, Pieper RO. (2005). IGFBP-3 generation tests in the diagnosis of growth hormone axis Akt activation suppresses Chk2-mediated, methylating agent- disturbances in children with beta-thalassaemia. Clin Endocrinol induced G2 arrest and protects from temozolomide-induced mitotic (Oxf) 54: 253–259. catastrophe and cellular senescence. Cancer Res 65: 4861–4869. Corcoran RB, Bachar Raveh T, Barakat MT, Lee EY, Scott MP. Kandel ES, Skeen J, Majewski N, Di Cristofano A, Pandolfi PP, (2008). Insulin-like growth factor 2 is required for progression to Feliciano CS et al. (2002). Activation of Akt/protein kinase B advanced medulloblastoma in patched1 heterozygous mice. Cancer overcomes a G(2)/m cell cycle checkpoint induced by DNA damage. Res 68: 8788–8795. Mol Cell Biol 22: 7831–7841. Dahmane N, Ruiz i Altaba A. (1999). Sonic hedgehog regulates Kenney AM, Rowitch DH. (2000). Sonic hedgehog promotes G(1) the growth and patterning of the cerebellum. Development 126: cyclin expression and sustained cell cycle progression in mammalian 3089–3100. neuronal precursors. Mol Cell Biol 20: 9055–9067. Dudek H, Datta SR, Franke TF, Birnbaum MJ, Yao R, Cooper GM Kenney AM, Widlund HR, Rowitch DH. (2004). Hedgehog and PI-3 et al. (1997). Regulation of neuronal survival by the serine– kinase signaling converge on Nmyc1 to promote cell cycle threonine protein kinase Akt. Science 275: 661–665. progression in cerebellar neuronal precursors. Development 131: Duronio V. (2008). The life of a cell: apoptosis regulation by the 217–228. PI3K/PKB pathway. Biochem J 415: 333–344. Kool M, Koster J, Bunt J, Hasselt NE, Lakeman A, van Sluis P et al. Eberhart CG. (2008). Even cancers want commitment: lineage identity (2008). Integrated genomics identifies five medulloblastoma sub- and medulloblastoma formation. Cancer Cell 14: 105–107. types with distinct genetic profiles, pathway signatures and Fernandez-L A, Kenney AM. (2010). The Hippo in the room: a new clinicopathological features. PLoS One 3: e3088. look at a key pathway in cell growth and transformation. Cell Cycle Leonard JM, Ye H, Wetmore C, Karnitz LM. (2008). Sonic Hedgehog 9: 2292–2299. signaling impairs ionizing radiation-induced checkpoint activation Fernandez-L A, Northcott PA, Dalton J, Fraga C, Ellison D, Angers S and induces genomic instability. J Cell Biol 183: 385–391. et al. (2009). YAP1 is amplified and up-regulated in hedgehog- Mueller S, Chang S. (2009). Pediatric brain tumors: current treatment associated medulloblastomas and mediates Sonic hedgehog-driven strategies and future therapeutic approaches. Neurotherapeutics 6: neural precursor proliferation. Genes Dev 23: 2729–2741. 570–586. Frappart PO, Lee Y, Russell HR, Chalhoub N, Wang YD, Orii KE Nahle Z, Hsieh M, Pietka T, Coburn CT, Grimaldi PA, Zhang MQ et al. (2009). Recurrent genomic alterations characterize medullo- et al. (2008). CD36-dependent regulation of muscle FoxO1 and blastoma arising from DNA double-strand break repair deficiency. PDK4 in the PPAR delta/beta-mediated adaptation to metabolic Proc Natl Acad Sci USA 106: 1880–1885. stress. J Biol Chem 283: 14317–14326. Goodrich LV, Milenkovic L, Higgins KM, Scott MP. (1997). Altered Negrini S, Gorgoulis VG, Halazonetis TD. (2010). Genomic insta- neural cell fates and medulloblastoma in mouse patched mutants. bility—an evolving hallmark of cancer. Nat Rev Mol Cell Biol 11: Science 277: 1109–1113. 220–228. Hahn H, Wojnowski L, Specht K, Kappler R, Calzada-Wack J, Northcott PA, Fernandez LA, Hagan JP, Ellison DW, Grajkowska W, Potter D et al. (2000). Patched target Igf2 is indispensable for Gillespie Y et al. (2009a). The miR-17/92 polycistron is up-regulated

Oncogene YAP promotes radioresistance in medulloblastoma A Fernandez-L et al 1937 in sonic hedgehog-driven medulloblastomas and induced by N-myc Reinhardt HC, Yaffe MB. (2009). Kinases that control the cell cycle in in sonic hedgehog-treated cerebellar neural precursors. Cancer Res response to DNA damage: Chk1, Chk2, and MK2. Curr Opin Cell 69: 3249–3255. Biol 21: 245–255. Northcott PA, Nakahara Y, Wu X, Feuk L, Ellison DW, Croul S et al. Ruiz i Altaba A. (1999). Gli proteins encode context-dependent (2009b). Multiple recurrent genetic events converge on control positive and negative functions: implications for development and of histone lysine methylation in medulloblastoma. Nat Genet 41: disease. Development 126: 3205–3216. 465–472. Shtivelman E, Sussman J, Stokoe D. (2002). A role for PI 3-kinase Northcott PA, Korshunov A, Witt H, Hielscher T, Eberhart CG, and PKB activity in the G2/M phase of the cell cycle. Curr Biol 12: Mack S et al. (2011). Medulloblastoma comprises four distinct 919–924. molecular variants. J Clin Oncol 29: 1408–1414. Sieber OM, Heinimann K, Tomlinson IP. (2003). Genomic instabil- Oliver TG, Grasfeder LL, Carroll AL, Kaiser C, Gillingham CL, Lin ity—the engine of tumorigenesis? Nat Rev Cancer 3: 701–708. SM et al. (2003). Transcriptional proﬁling of the Sonic hedgehog Tanori M, Mancuso M, Pasquali E, Leonardi S, Rebessi S, response: a critical role for N-myc in proliferation of neuronal Di Majo V et al. (2008). PARP-1 cooperates with Ptc1 to suppress precursors. Proc Natl Acad Sci USA 100: 7331–7336. medulloblastoma and basal cell carcinoma. Carcinogenesis 29: Overholtzer M, Zhang J, Smolen GA, Muir B, Li W, Sgroi DC et al. 1911–1919. (2006). Transforming properties of YAP, a candidate oncogene on Tanori M, Santone M, Mancuso M, Pasquali E, Leonardi S, the chromosome 11q22 amplicon. Proc Natl Acad Sci USA 103: Di Majo V et al. (2010). Developmental and oncogenic effects of 12405–12410. insulin-like growth factor-I in Ptc1+/À mouse cerebellum. Mol Parathath SR, Mainwaring LA, Fernandez LA, Campbell DO, Cancer 9: 53. Kenney AM. (2008). Insulin receptor substrate 1 is an effector of Wallace VA. (1999). Purkinje-cell-derived Sonic hedgehog regulates sonic hedgehog mitogenic signaling in cerebellar neural precursors. granule neuron precursor cell proliferation in the developing mouse Development 135: 3291–3300. cerebellum. Curr Biol 9: 445–448. Pazzaglia S, Tanori M, Mancuso M, Rebessi S, Leonardi S, Di Majo V Wechsler-Reya RJ, Scott MP. (1999). Control of neuronal precursor et al. (2006). Linking DNA damage to medulloblastoma tumori- proliferation in the cerebellum by Sonic Hedgehog. Neuron 22: genesis in patched heterozygous knockout mice. Oncogene 25: 103–114. 1165–1173. Xu N, Hegarat N, Black EJ, Scott MT, Hochegger H, Gillespie DA. Plo I, Laulier C, Gauthier L, Lebrun F, Calvo F, Lopez BS. (2008). (2010). Akt/PKB suppresses DNA damage processing and check- AKT1 inhibits homologous recombination by inducing cytoplasmic point activation in late G2. J Cell Biol 190: 297–305. retention of BRCA1 and RAD51. Cancer Res 68: 9404–9412. Yuan M, Tomlinson V, Lara R, Holliday D, Chelala C, Harada T Pomeroy SL, Tamayo P, Gaasenbeek M, Sturla LM, Angelo M, et al. (2008). Yes-associated protein (YAP) functions as a tumor McLaughlin ME et al. (2002). Prediction of central nervous system suppressor in breast. Cell Death Differ 15: 1752–1759. embryonal tumour outcome based on gene expression. Nature 415: Zeng Q, Hong W. (2008). The emerging role of the hippo pathway in 436–442. cell contact inhibition, organ size control, and cancer development Rao G, Pedone CA, Del Valle L, Reiss K, Holland EC, Fults DW. in mammals. Cancer Cell 13: 188–192. (2004). Sonic hedgehog and insulin-like growth factor signaling Zhao B, Ye X, Yu J, Li L, Li W, Li S et al. (2008). TEAD mediates synergize to induce medulloblastoma formation from nestin- YAP-dependent gene induction and growth control. Genes Dev 22: expressing neural progenitors in mice. Oncogene 23: 6156–6162. 1962–1971.

Supplementary Information accompanies the paper on the Oncogene website (http://www.nature.com/onc)

Oncogene